Machine perfusion system

The machine perfusion (MPS) system was designed to perfuse the heart at varying flow rates. The UW Belzer MPS solution (Belzer MPS®, Bridge to Life Ltd., Columbia, SC) was placed into a preservation solution reservoir and pushed into the system using a roller pump to deliver low volumes of solution (Fisher Scientific, Hampton, NH). The composition of the UW Belzer MPS solution is reported in Table 1. This is an isotonic solution that has a supraphysiological potassium concentration that maintains heart in diastolic arrest. It has been designed to reduce the cellular metabolic demand, reduce edema and maintain cell wall integrity, sustain basal metabolism and scavenge free radicals. The flow rate was measured using an in-line flow probe (AD instrument), following passage through an oxygenator (adapted from a Hemofilter D150, Medica, Monza, Italy). Placement of a cooling/heating chamber in front of the heart cannula enabled collection of sample solution with a 3-way stop-cock before entering the heart (input solution). The aortic cannula perfused the heart while the inferior vena cava cannula collected the perfusate exiting the coronary sinus (output solution). A simplified schema of the experimental protocol and MPS is represented in Fig. 1.

Composition of the UW Belzer MPS solution

Simplified schema of the experimental design and protocol. Panel a. Experimental protocol employed to determine the ex-vivo critical flow index (D_CRIT) of isolated rat hearts according to changes in the composition and temperature of the perfusate. Panel b. The machine perfusion apparatus is primed with oxygenated UW Belzer MPS solution at the desired temperature by using a recirculating temperature controller and glass heat exchanger coil. Hearts are attached to the apparatus, and antegrade coronary flow is initiated via the aortic cannula at a controlled flow index (FI) by using a roller pump. Oxygen transfer to the perfusate is maintained by passing the solution through a pediatric hollow fiber hemofilter with a continuous oxygen sweep delivered across the outer compartment. Flow rate, flow pressure, and temperature of perfusate are monitored continuously. Input (aortic) and output solution (inferior vena cava) samples are collected serially to assess variables of interest

The partial pressure of O₂ in the input MPS solution (PaO₂) was above 700 mmHg at 15 °C (mean PaO₂ 724 ± 24), above 650 mmHg at 22 °C (mean PaO₂ 766 ± 27), and above 450 mmHg at 37 °C (mean PaO₂ 526 ± 55). The differences in these partial pressure values are due to the effect of temperature on oxygen solubility in a non-blood based solution. Flow values were expressed as FI as a standard measure for organ perfusion. All hearts in this study were set at an initial FI of at least 380 mL/min/100 g, equivalent to a flow rate of at least 3.8 ml/g/min. Input and output perfusate samples were collected 10 min from the beginning of the perfusion, using ice-cold 1 mL syringes. The syringes were capped with parafilm, placed on ice and rapidly analyzed (< 1 min) with the ABL-800 blood gas analyzer (Radiometer, Copenhagen, Denmark). Rigorous steps such as, pre-cooling, sealing and rapid reading, were necessary to ensure reliable readings from the samples. The perfusate flow rate was decreased by approximately 40–60 mL/min/100 g, or 0.4–0.6 ml/g/min, every 10 min after each measurement, until reaching the lowest perfusion rate allowed by the pump (30–40 mL/min/100 g, or 0.3–0.4 ml/g/min).

Flow rate and temperature of perfusate were monitored continuously with a Power Lab data acquisition system (AD Instruments, Denver, CO). MVO₂ values were calculated at each FI using the Fick equation [(CaO₂-CvO₂) x flow / heart weight], where CaO₂ and CvO₂ represent the concentration of O₂ in the input and output solution, respectively. MVO₂ and FI values were plotted in a graph. Initiation of the anaerobic phase is represented by the corresponding FI (D_CRIT) when MVO₂ becomes flow limited. The exact value of the D_CRIT was determined from the best fit lines of the points in the plateau phase (aerobic, above the D_CRIT) and the line derived in the flow limited part of the curve (anaerobic, below the D_CRIT) (Fig. 2) [10]. The FI corresponding to the point when MVO₂ drops to become flow limited represents D_CRIT. However, when the MVO₂ always remained flow limited (never reached a stable plateau phase) we could not determine the D_CRIT for the FI used.

Illustration depicting the D_CRIT in ex-vivo heart perfusion models. The point above which oxygen extraction is flow independent is D_CRIT. The isolated heart could not be efficiently preserved below this critical flow index since it will become hypoxic/ischemic (below “anaerobic threshold”) leading to lactic acidosis, impaired microcirculation and tissue damage. The “plateau value” (maximum MVO₂ in relation with the temperature of the perfusate) is achieved once the critical flow value is exceeded ensuring optimal perfusion conditions