If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Reprint requests: Daisuke Sakota, PhD, Health and Medical Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), 1-2-1 Namiki, Tsukuba, Ibaraki 3058564, Japan. Telephone: +81-29-861-2332.
Department of Advanced Surgical Technology Research and Development, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Tokyo, Japan
Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OhioDepartment of Inflammation and Immunology, Lerner Research Institute, Cleveland Clinic, Cleveland, OhioTransplant Center, Cleveland Clinic, Cleveland, Ohio
Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OhioDepartment of Inflammation and Immunology, Lerner Research Institute, Cleveland Clinic, Cleveland, OhioTransplant Center, Cleveland Clinic, Cleveland, Ohio
Department of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, OhioDepartment of Inflammation and Immunology, Lerner Research Institute, Cleveland Clinic, Cleveland, OhioTransplant Center, Cleveland Clinic, Cleveland, Ohio
For normothermic ex vivo heart perfusion (EVHP), a resting mode and working mode have been proposed. We newly developed a left ventricular assist device (LVAD) mode that supports heart contraction by co-pulse synchronized LVAD.
Methods
Following resting mode during time 0 to 1 hour, pig hearts (n = 18) were perfused in either resting, working, or LVAD mode during time 1 to 5 hour, and then myocardial function was evaluated in working mode at 6 hour. The preservation ratio was defined as the myocardial mechanical function at 330 minute divided by the function at 75 minute. In LVAD mode, LVAD unloaded the pressure and the volume in the left ventricle in the systolic phase.
Results
The LVAD group was significantly associated with higher preservation ratios in cardiac output (resting, 33 ± 3; working, 35 ± 5; LVAD, 76% ± 5%; p < 0.001), stroke work, dP/dt maximum, and dP/dt minimum compared with the other groups. Glucose consumption was significantly reduced in the resting group. The LVAD group was significantly associated with higher myocardial oxygen consumption (resting, 2.2 ± 0.3; working; 4.6 ± 0.5; LVAD, 6.1 ± 0.5 mL O2/min/100 g, p < 0.001) and higher adenosine triphosphate (ATP) levels (resting, 1.1 ± 0.1; working, 0.7 ± 0.1; LVAD, 1.6 ± 0.2 μmol/g, p = 0.001) compared with the others.
Conclusion
These data suggest that myocardial mechanical function was better preserved in LVAD mode than in resting and working modes. Although our data suggested similar glycolysis activity in the LVAD and working groups, the higher final ATP in the LVAD group might be explained by reduced external work in LVAD.
Normothermic ex vivo heart perfusion (EVHP) is a clinically applied technique that preserves a donor heart by providing energy and nutrients to it in a beating, semi-physiological condition, thus minimizing cold ischemic time.
A prospective multicenter randomized study demonstrated that 30-day graft survival using the TransMedics Organ Care System (OCS) Heart was comparable to static cold storage.
Ex-vivo perfusion of donor hearts for human heart transplantation (PRODEED II): a prospective, open-label, multicentre, randomized non-inferiority trial.
Preservation time (cold ischemia + perfusion) in the OCS Heart group was longer than that of the control (324 vs 195 minute). Resting mode EVHP may be thus considered superior to static cold storage with regard to the longer preservation time.
OCS Heart as a resting mode lacks myocardial mechanical functional assessment.
Ex-vivo perfusion of donor hearts for human heart transplantation (PRODEED II): a prospective, open-label, multicentre, randomized non-inferiority trial.
To overcome these limitations, a working mode was developed. A working mode, in which perfusate is provided to the left heart, allows mechanical assessment of left heart function during EVHP. Most systems use a 2-chamber working mode in which perfusate is provided to the left atrium (LA) and ejected from the left ventricle (LV).
Recently, Hatami et al. reported that 12 hour of the working mode results in better functional preservation and a lower rate of apoptosis than the resting mode.
In spite of there being no significant differences in myocardial oxygen consumption (MVO2) and free fatty acid (FFA) levels, glucose utilization in the working mode is increased. However, the mechanism of these metabolic characteristics in both modes remains unknown.
The left ventricular assist device (LVAD) has been clinically used as a therapeutic option in patients with advanced heart failure, either as a bridge to transplantation,
However, the effect of LVAD on ex vivo donor heart preservation has not been investigated. Herein, we developed a new LVAD mode: a novel working mode EVHP system that supports heart contraction by co-pulse synchronized LVAD. The developed LVAD mode expresses the preload in the same way as the conventional working mode in diastolic phase, but unloads the pressure and the volume in the LV in the systolic phase. The effects of preload on the heart can be divided into the following during diastole and systole: (1) to fill LV with perfusate from LA during diastole, and (2) to eject the perfusate during systole. Here, the presence or absence of the first role is described by “LA+” (LA loading) or “LA−“ (LA unloading), respectively. As for the second role, it is described by “LV+” or “LV−“. Resting mode performs LA− and LV−, whereas working mode performs LA+ and LV+. However, the second role during systole might be exempted in the LVAD mode by assisting contraction of the heart. Therefore, in this study, the LVAD mode is defined as LA+, LV−. We hypothesized that this LVAD mode might preserve myocardial function better than conventional resting and working modes. The purposes of this study were to compare myocardial mechanical function and metabolic parameters between resting and working modes, and to compare the LVAD mode against alternative modes.
Materials and methods
Study design
Male pigs, crossbred between Landrace and Large White pigs (76 ± 2 kg, n = 18) were randomly divided into 3 experimental groups (Figure 1): a resting group (LA−, LV−) as a control, a working group (LA+, LV+), and an LVAD group (LA+, LV−) (n = 6, each). At 0 minute, resting mode was started and switched to working mode at 60 minute, and baseline myocardial function was evaluated at 75 minute in all groups. Then, the heart perfusion mode was changed to each group's mode and maintained until 300 minute. At 300 minute, the working mode was initiated in the resting and LVAD groups, while there was no change in the working group. Myocardial function was evaluated in all groups at 330 and 360 minute. The protocol was approved by the Institutional Animal Care and Use Committees of the National Institute of Advanced Industrial Science and Technology and Tokyo Medical and Dental University.
Figure 1Study design. Following resting mode for 0 to 60 minute, baseline myocardial mechanical function was measured at 75 minute in working mode in all groups. Then, resting mode was maintained in the resting group until 300 minute. In the LVAD group, LVAD mode was kept from time 75 to 300 minute. At 330 and 360 minute, myocardial mechanical function was measured in working mode in all groups.
Details of EVHP are described in Figure 2 and Appendix 2. Briefly, a combination of 1,600 mL autologous whole blood and 2,000 mL STEEN solution was used. The circuit consisted of 3 centrifugal blood pumps, a membrane oxygenator, a reservoir, a compliance chamber with Windkessel, and an organ chamber. Mean left atrium pressure (LAP) was maintained at 6 to 8 mm Hg in the working and LVAD modes. Two types of evaluation were performed: a regular afterload of 40 mm Hg at 330 minute and an increased afterload of 60 mm Hg at 360 minute. Heart weight was measured at 360 minute.
Figure 2Ex vivo heart perfusion circuit. The height of the compliance chamber of the aortic root corresponding to afterload was adjusted to 40 mm Hg. In the LVAD group, the LV inflow cannula (black) was inserted into the LV, and all lines were opened. The EVHP apparatus consisted of 3 centrifugal blood pumps, a membrane oxygenator, a reservoir, a compliance chamber with Windkessel to provide vascular impedance in the aortic root, and an organ chamber. All centrifugal blood pumps including the LVAD pump were automatically controlled by developed software (LabVIEW version 2018; National Instruments, Tokyo, Japan). In the evaluation at 360 minute, the height of the compliance chamber was changed to 60 mm Hg.
At 75 minute, a rotary blood pump (Gyro pump C1E3; Kyocera Corp., Kyoto, Japan) was started in the LVAD group. The monitored value of the end LVP became a trigger for a boost in LVAD rotational speed. At that moment, the pulsatile flow by LVAD was generated. A system operator matched up the peak of the first-order derivative of the LVAD flow rate (dF/dt maximum) with that of LVP (dP/dt maximum) as shown in Figure 3A. To evaluate its co-pulse synchronization performance, phase differences of dF/dt maximum and dP/dt maximum were calculated (Figure 3B). The second-order derivative of LVP (d2P/t2) maximum was set as time 0, indicating the approximate start of left ventricular contraction. The peak difference between these 2 distributions was 10 ms. The full width at half-maximum of the distribution of dF/dt maximum was 120 ms. The synchronized ratio between dF/dt maximum and d2P/dt2 maximum within 100 ms was 76%.
Figure 3(A). Typical waveform of left ventricular pressure (LVP), first-order derivative of LVP (dP/dt), second-order derivative of LVP (d2P/dt2), LVAD flow rate (F), and first-order derivative of F (dF/dt). (B). Histogram of the phase difference of dF/dt maximum and dP/dt maximum for d2P/dt2 maximum. The total number of heart beats was 23,490 ± 992 (n = 6) during LVAD mode.
Myocardial function variables, including cardiac output (CO), stroke work (SW), dP/dt maximum, dP/dt minimum, and LVP maximum, were assessed at 75, 120, 180, 240, 300, 330, and 360 minute. Preserved myocardial mechanical function was defined as myocardial function at 330 or 360 minute divided by the function at 75 minute. The CO was determined by measuring the flow rate through the LA line.
During LVAD mode, isolated CO (iCO) was calculated as CO – the flow rate through the LVAD line. The SW was calculated as the product of LV developed pressure: mean LVP – mean LAP, and stroke volume: CO/heart rate.
The perfusate samples were collected from the membrane and pulmonary artery cannula every hour. Blood gas analysis was performed (ABL80 FLEX; Radiometer KK, Tokyo, Japan). MVO2 was calculated by multiplying the coronary blood flow by the arterial-venous difference in oxygen content.
The pulmonary artery blood flow rate was regarded as equal to the coronary blood flow. In addition, FFA, glucose, lactate, and pyruvate levels were measured. The lactate/pyruvate (L/P) ratio was calculated to evaluate anaerobic metabolism.
As a sham group, baseline ATP levels were measured using fresh hearts procured following minimal cold ischemic time (n = 3).
Statistical analysis
Statistical analysis was performed using JMP version 15.2.1 (SAS Institute Inc., Cary, NC). Data were expressed as mean ± standard error. First, 3 groups were compared using analysis of variance (ANOVA). When significant differences were identified, 2 sample Student's t-test in all pairs of groups. Then, Holm correction was applied to adjust multiple comparisons. A value of p < 0.05 was considered statistically significant.
Results
Resting group vs working group
There was no significant difference in heart rate between the resting and working groups throughout the experiment (Figure 4A). Trends of CO, SW, dP/dt maximum, dP/dt minimum, and LVP maximum in the working group declined at 75 to 300 minute in a time-dependent fashion (Figure 4B-F). In the resting group, no myocardial functions were available at 120 to 300 minute because of the non-working state. All myocardial mechanical functions at 360 minute increased compared with 330 minute in both groups (Figure 4, Table 1). Estimated coronary blood flow at 360 min was significantly higher than at 330 minute, whereas coronary blood flow did not significantly differ between resting and working groups at 330 and 360 minute. The resting group demonstrated significantly lower dP/dt minimum at 330 minute and higher CO at 360 minute than the working group (Table 1). In the metabolic analysis, MVO2 in the resting group was significantly lower than in the working group at 180 to 360 minute (Figure 5A). Lactate levels in both groups rapidly increased at 60 to 120 minute (Figure 5B). At 180 to 300 minute, the lactate level in the resting group was stable, while that in the working mode increased with a significant difference. The pyruvate level in the resting group showed milder increase and the level was significantly lower than that in the working group at 360 minute (Figure 5C). The L/P ratio in the resting group was significantly lower than that in the working group at 120 to 300 minute (Figure 5D). The glucose level in the working group was significantly lower than that in the resting group at 300 minute (Figure 5E). There was no difference in FFA between the 2 groups at 120 to 300 minute (Figure 5F). At 360 minute, ATP level of working group was significantly lower than in the sham group (working, 0.7 ± 0.1 vs sham, 1.7 ± 0.2 μmol/g, p = 0.002, Figure 6). Furthermore, the ATP level of the working group was significantly lower than that in the resting group (resting, 1.1 ± 0.1 μmol/g, p = 0.044). Heart weight changes in both groups were similar (resting, 4 ± 4% vs working, 6 ± 4%).
Figure 4Myocardial mechanical function in EVHP.(A). Heart rate. (B). Cardiac output. (C). Stroke work. (D). dP/dt maximum. E. dP/dt minimum. (E). LVP maximum. * p < 0.05: resting vs working groups. # p < 0.05: resting vs LVAD groups. † p < 0.05: working vs LVAD groups.
Abbreviations: CO, cardiac output; SW, stroke work; LVP, left ventricular pressure; dP/dt, the first-order derivative of LVP.
Preserved myocardial mechanical function was defined as myocardial function at 330 or 360 min/the function at 75 minute. The height of compliance chamber of aortic root was adjusted to 40 mm Hg. In the evaluation at 360 minute, the height was changed to 60 mm Hg.
Figure 5Metabolic parameters in EVHP. (A) Myocardial oxygen consumption. (B). Lactate. (C). Pyruvate. (D). Lactate/pyruvate (L/P) ratio. The baseline value was at 60 minute (resting, 21 ± 3; working, 16 ± 1; LVAD, 14 ± 0.3 mg/dL). (E). Glucose. The baseline value was at 60 minute (resting, 155 ± 19; working, 137 ± 7; LVAD, 134 ± 6 mg/dL). F. FFA, free fatty acids. The baseline value was at 60 minteu (resting, 3372 ± 92; working, 3710 ± 89; LVAD, 3817 ± 91 mg/dL). * p < 0.05: resting vs working groups. # p < 0.05: resting vs LVAD groups. † p < 0.05: working vs LVAD groups.
Figure 6ATP levels at the end of EVHP. Tissue ATP in the resting, working, or LVAD group was obtained at 360 minute perfusion following the manner written in Appendix 3. In the sham group (n = 3), hearts were explanted with the same surgical technique as the other groups. After the completion of cardioplegia infusion using St. Thomas Hospital Solution at 4°C, the heart was extracted and the tissue sampling was completed after 10 minute of cold ischemic time.
In the LVAD group, the heart rate was not significantly different from the other groups (Figure 4A). During the support of LVAD at 120 to 300 minute, the total CO—which is equal to iCO plus LVAD flow—and SW of the LVAD group improved, whereas those of the working group declined (Figure 4B, C). The average support ratio, defined as LVAD flow/total CO × 100, was 71 ± 2% (Figure 4B). At 330 min, the CO, SW, dP/dt maximum, dP/dt minimum, LVP maximum, and preservation ratio of the LVAD group were significantly higher compared with the other groups (Figure 4B-4D, Table 1). The trends of metabolic variables (lactate, pyruvate, L/P ratio, and glucose) in the LVAD group were similar to those of the working group at 120 to 300 minute (Figure 5B-E). For FFA levels, there were no significant differences among the 3 groups (Figure 5F). At 360 minute, ATP levels in the LVAD group were significantly higher than in the working group (LVAD, 1.6 ± 0.2 vs working, 0.7 ± 0.1 μmol/g, p = 0.007, Figure 6), and the level was similar to that of the sham group (sham, 1.7 ± 0.2 μmol/g, p = 0.743). There were no significant differences among the 3 groups in change in heart weight (Table 1).
Discussion
In this study, we investigated the utility of a novel LVAD mode that supports left ventricular contraction in the conventional working mode during EVHP. Myocardial function and biomarkers were evaluated among resting, working, and LVAD groups during 6 hour of pig heart EVHP. The LVAD group was significantly associated with better myocardial function, higher MVO2, and higher ATP levels than the resting or working groups, though other biomarkers (lactate, pyruvate, glucose, and FFA) of the LVAD mode were similar to those of the working group. Therefore, the LVAD mode is expected to have better preservation performance than conventional EVHP.
In the present study, myocardial function was evaluated at 2 time points, 330 and 360 minute, under working mode. The afterload at 330 minute was the same as the baseline assessment at 75 minute, whereas the afterload at 360 minute was increased. Then, higher coronary flow was observed at 360 minute compared with 330 minute. The improving trend of myocardial function at 360 minute could be explained by the increase in coronary flow.
Sarnoff et al reported that aortic pressure significantly increases when total coronary flow is elevated in a dog EVHP model.
In the comparison between resting and working groups, the working group was significantly associated with a lower glucose level, higher lactate level, and higher pyruvate level than the resting group, suggesting that glycolysis was enhanced in the working group. Furthermore, the higher L/P ratio in the working group might indicate that anaerobic metabolism was augmented in the working group. In contrast to glycolysis, there was no difference in FFA metabolism between the resting and working groups. These results are consistent with a report by Hatami et al. in which glucose consumption in the working group was significantly higher than that in the resting group, and there was no difference in FFA consumption between the 2 groups.
The metabolism of glycolysis and FFA might progress in an independent fashion during EVHP. In addition, the lower MVO2 of the resting group in our study suggests that its myocardial metabolism was suppressed. In the resting group, the CO was zero at 120 to 300 minute because of no preloading. It has been reported that the amount of MVO2 increases in proportion to the contractility of the heart, whereas only 10% to 20% of total MVO2 is utilized for maintaining basal metabolism.
Therefore, based on the low MVO2 in the resting group, lower ATP production was expected. The final ATP level is decided by the balance of ATP production and consumption, and the ATP level at 360 minute in the resting group was significantly higher compared to the working group. Significantly higher external work by the working group (vs no work by the resting group) suggests that high amounts of ATP were consumed. In this study, the resting group was significantly associated with worse diastolic function (dP/dt minimum) than the working group. In contrast, systolic function (CO) of the resting group was better than that of the working group. These results suggest that diastolic failure was enhanced under the no preload state due to a lack of LV dilatation. The better systolic function of the resting group can be explained by the higher ATP level. In addition, significantly lower ATP level was indicated in the working group. Therefore, these mechanical and metabolic data, especially the CO and ATP values, suggest that resting mode is better for perfusion than working mode for the purpose of “preservation.”
When comparing resting and working modes, Hatami et al. demonstrated that the working mode was significantly associated with better CO, SW, and dP/dt minimum at 5 and 11 hour compared to the resting mode.
Their metabolic data showed no difference in MVO2 between the 2 modes, whereas the trend of glucose and FFA levels was similar to our data. The discrepancy in myocardial function and MVO2 from our data might be explained by the difference in afterload during perfusion: a centrifugal blood pump on the aortic root was utilized by Hatami et al., whereas a compliance chamber was used in our study.
Interestingly, the LVAD group demonstrated better myocardial function and higher ATP levels than the other 2 groups, while there was no difference in metabolic profiles of glycolysis between the LVAD and working groups. These results might be related to the 2 roles of preload of LA+ and LV−. The working group performed both roles (LA+, LV+), whereas the second role was exempted in the LVAD group because of LVAD support. The first role in diastole of “LA+” is thought to have the effect of increasing MVO2, resulting in higher ATP production. The oxygen demand is related to ventricular wall tension triggered by the first role.
Based on the significantly lower iCO in the LVAD group compared to CO in the working group, the external work of the LVAD group was apparently extremely limited. Therefore, ATP consumption might be lower in the LVAD group by “LV−”. In addition, the final ATP level in the LVAD group remained the same as in the sham group and was the highest among all groups. These results imply that it is important to maintain a balance between ATP production and consumption in terms of heart preservation by EVHP. Preload has the benefit of facilitating metabolic activity of the heart (ATP production), but it has the detriment of making the heart do external work and increasing ATP consumption. However, ATP consumption can be reduced by assisting contraction of the heart.
This study has several limitations. First, there was a small number of animals in each group. Second, the study design included several discrepancies between clinical applications of these modes. The working and LVAD modes were only possible when the heart was restored to normothermic temperature and beating. Third, this study included no heart transplantation model, so a further study is necessary. However, Gellner et al. reported that there is a significant correlation between parameters during pig heart working mode EVHP and post-transplant cardiac function.
Fourth, 100% synchronization was not achieved due to arrhythmia, and the support ratio of the LVAP pump was approximately 72% due to limitations of LVAD pump function. These limitations might have made ATP levels in Figure 6 much wider than those of other groups. Improving synchronization and support ratios might further result in better preservation performance in LVAD mode. Fifth, it might be necessary to simplify LVAD mode in future clinical use.
In conclusion, the LVAD group was significantly associated with better cardiac mechanical function, higher MVO2, and higher ATP levels than the other groups. These results suggest that myocardial function was better preserved in the LVAD mode compared to the resting and working modes during EVHP. Reduced external work during LVAD mode might lead to reduced ATP consumption and higher final ATP levels at the end of EVHP.
Author contributions
D.S.: study design, experiments, data analysis, manuscript preparation. R.K.: experiments, data analysis, manuscript revision. E.N.: experiments, manuscript revision. K.O.: experiments, data analysis, manuscript revision. T.T.: experiments. H.A.: study design, manuscript revision. I.S.: data analysis, manuscript revision. K.R.M.: study design, manuscript revision. T.O.: study design, data analysis, manuscript revision.
Disclosure Statement
The authors have no conflicts of interest to disclose.
This study was financially supported by the AIST program for young researchers (AIST EDGE Runners, Recipient: Daisuke Sakota) and JSPS KAKENHI Grant Number JP19H03723. We thank Dr. Yoshifumi Itoda and Dr. Ko Sakatsume for assistance with the EVHP setup and Dr. Darren H. Freed for instruction on EVHP system development.
Appendix 1: Heart procurement and preparation
Tracheal intubation followed subcutaneous injection of 2 mg/kg xylazine and 20 mg/kg ketamine with isoflurane of 1.5% to 3.0% started. The electrocardiogram, oxygen saturation, and arterial or aortic pressure were monitored. A standard median sternotomy was performed. After heparinization, a drainage catheter into the superior vena cava and an aortic root cannula were placed. Prior to exsanguination, 1,600 mL of whole blood was collected from the venous catheter to a bottle containing heparin 10,000 IU and imipenem 100 mg. Then, the superior vena cava was ligated following removal of the venous catheter, the inferior vena cava was amputated to drain the venous blood, and the ascending aorta was cross-clamped. Immediately thereafter, 2,000 mL of cardioplegia, St. Thomas Hospital Solution (Miotecter; Mochida Pharmaceutical Co. Tokyo, Japan) at 4°C, was delivered ante grade through the aortic root cannula. To prevent left ventricle distension, the right upper pulmonary vein was dissected in advance and cut down immediately after heart arrest. Simultaneously, crushed ice was bedded into the pericardial sac to start cold storage. After the completion of cardioplegia infusion, the heart was extracted and weighed. Then, each heart was soaked in 4°C saline for 30 minute, and the following cannulations were performed. First, a specially developed LA cannula was placed in the LA (Figure 2). Second, an LV cannula was inserted from the side branch of the LA cannula into the LV. Third, an aortic cannula was secured in the aorta. A thermometer probe was inserted from the superior vena cava to the right atrium, and the superior vena cava was ligated. A purse string suture was placed in the inferior vena cava with ligation after attaining a stable heartbeat to collect the cardiac venous blood from the pulmonary artery trunk. An epicardial bipolar pacing lead was placed on the right ventricle to control heart rate.
All animals received humane care in compliance with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources and published by the National Institutes of Health (NIH Publication No. 86-23, revised 1996).
Appendix 2: Normothermic ex vivo heart perfusion
The EVHP system was primed with a combination of 1,600 mL autologous whole blood and 2,000 mL STEEN solution. The perfusate temperature was set at 22°C during the priming.
The heart was placed on a silicon hammock sheet in an organ chamber. At 0 minute, the aortic cannula was connected to the aortic line, and resting mode was started. For afterload control, the height was strictly adjusted so that the height from the aortic valve position to the level of the compliance chamber was 40 mm Hg. Throughout EVHP, the following medications were continuously infused: glucose (0.42 g/h, OTSUKA GLUCOSE INJECTION; Otsuka Pharmaceutical Co. Ltd., Tokyo, Japan), insulin (3 U/h, NovoRapid Injection; Novo Nordisk Pharma Ltd., Tokyo, Japan), and dobutamine (5 µg/min, DOBUTAMINE injection 100 mg; Pfizer Inc., NY). PaO2 of the aortic line was kept at approximately 300 mm Hg: a gas mixture of 38% of O2, 60% of N2, and 2% of CO2 was delivered to a membrane oxygenator at a flow rate of 8 L/min. pH was maintained at 7.30 to 7.45. A pulmonary artery cannula was secured to the main pulmonary artery. A Millar catheter was positioned in LV through the apex to monitor LVP. At 30 min, perfusate temperature was set to 37°C. The heart was defibrillated with up to 20 J when needed. The heart was paced if the heart rate was < 90 beats/min. After the heart started to beat, the inferior vena cava was ligated, and then all outflow from the coronary sinus was collected to the pulmonary artery cannula. At 60 min, working mode was started. At 60 to 75 minute, mean LAP gradually increased and was maintained at 6 to 8 mm Hg. For preload control, the mean LAP per second was set as the target value, and the pump speed was constantly feedback-controlled to automatically reach that pressure. In the resting group, the LA cannula was disconnected from the LA line, and then the heart was transitioned to an unloaded state. At 300 minute, the LA cannula was reconnected to the LA line, and working mode was started. In the working group, the preload control was maintained during time 60 to 360 minute. Baseline myocardial mechanical function was evaluated at 75 minute. In all groups, the LV cannula was removed at 320 minute. In the evaluation at 360 minute, the height of the compliance chamber was changed to 60 mm Hg.
Appendix 3: Tissue sampling for ATP measurement
At 360 minute, all cannulas and probes except the aortic cannula were carefully removed from the heart in an organ chamber. Then, the inferior vena cava was opened again, and the aorta line was clamped. Immediately thereafter, 1,000 mL of cardioplegia, St. Thomas Hospital Solution at 4°C, was delivered antegrade through the aortic root. After completion, the heart was disconnected from the circuit. Then the heart weight was measured, and the heart weight gain between before and after EVHP was calculated to evaluate myocardial edema. The disconnected heart from the circuit was cut horizontally at 1-cm thickness at the center position. The sectioned LV tissue was immediately cryopreserved in liquid nitrogen and stored at −80°C. The tissue was then thawed and homogenized to measure ATP.
References
Vela MM
Sáez DG
Simon AR.
Current approaches in retrieval and heart preservation.
Ex-vivo perfusion of donor hearts for human heart transplantation (PRODEED II): a prospective, open-label, multicentre, randomized non-inferiority trial.
00029-3&id=gr1.jpg){kind=link}
00029-3&id=gr2.jpg){kind=link}
00029-3&id=gr3.jpg){kind=link}
00029-3&id=gr4.jpg){kind=link}
00029-3&id=gr5.jpg){kind=link}
00029-3&id=gr6.jpg){kind=link}