|
|
||||||||
Eur J Cardiothorac Surg 2001;20:187-194
© 2001 Elsevier Science NL
a Department of Surgery, University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK
b Comparative Biology Centre, University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne, NE2 4HH, UK
Received 10 October 2000; received in revised form 12 March 2001; accepted 30 March 2001.
Corresponding author. William Leech Centre, Freeman Hospital, High Heaton, Newcastle upon Tyne, NE7 7DN, UK. Tel.: +44-191-284-3111, ext. 26169; fax: +44-191-223-1152
e-mail: j.d.aitchison{at}ncl.ac.uk
| Abstract |
|---|
|
|
|---|
Key Words: Lung transplantation Non-heart-beating donor Isolated perfusion Animal
| 1. Introduction |
|---|
|
|
|---|
There are several successful European schemes for NHBD renal transplantation, including those in Maastricht [6], Leicester [7] and our own city [8]. Long-term results of NHBD renal transplantation are acceptable, but delayed function requiring renal support is a problem in over 60% of cases [9].
The lung may be more amenable to non-heart-beating donation than the kidney, by virtue of its structure. Tissue levels of high-energy nucleotide phosphates remain near normal for up to 4 h after the cessation of circulation, provided the lung remains inflated with oxygen [10]. Work with different animal models has consistently shown that lungs transplanted after 12 h of warm ischaemia perform satisfactorily [1113]. Inflation, with or without ventilation, with either oxygen or air, for a period of a few hours provides optimal storage conditions [14,15].
However, the selection of suitable lungs for transplantation remains a problem; currently only 2025% of brain-stem dead multiple organ donors have lungs suitable for transplantation due to trauma, infection or inflammatory changes [3]. The donor pulmonary venous oxygen tension remains our best marker for the prediction of satisfactory initial function post-transplantation [16]. In spite of this, poor post-transplantation function remains a major problem, with function significantly impaired in 1020% [17]. If, like the kidney, there is an increased incidence in poor organ function from NHBD, then an accurate method for the prediction of poor post-transplantation lung function is vital. Likewise, pre-existing poor pulmonary function in the donor prior to arrest would also need to be identified. Unlike current brain-stem dead but heart-beating donors, this assessment would need to be performed in the absence of respiration and circulation. Our aim, therefore, was to develop a ventilation and perfusion assessment technique for non-heart-beating donor lungs that was able to differentiate between those lungs with good and poor function post-transplantation.
| 2. Materials and methods |
|---|
|
|
|---|
The left single lung transplant model used in our centre has been previously published in detail [18], and the procedures described are modified from this technique. Landrace cross Yorkshire White juvenile pigs of 50 kg approximate weight were used for all experiments; group weights are shown in Table 1. Donor and recipient were inbred littermates and for ease of surgery, the larger animal was chosen as the recipient. Lungs were retrieved from control groups (Control1 and Control2,4) and after 1, 2 and 4 h from NHBD animals after hypoxic cardiac arrest (NHBD1, NHBD2, and NHBD4). A schematic diagram in Fig. 1 shows the stages for each group as described below. Two control groups were used; for the Control1 group, the left lung was subjected to ventilation and perfusion assessment and was subsequently transplanted. In view of the poor function post-transplantation of lungs previously assessed in this manner, for further experiments, the right lung was assessed with the left lung transplanted. Assessment results therefore required a different, right lung control group, namely Control2,4.
|
|
For control lung donors, Polystan 16 and 21 French gauge perfusion cannulae, respectively were inserted into the main pulmonary artery and left atrial appendage through purse-string sutures. Immediately prior to arrest, blood withdrawal was commenced via the inferior vena cava. Cardiac inflow was prevented by ligation of the superior and inferior vena cavae at their junctions with the right atrium. Aortic outflow was occluded by cross-clamping and both lungs were perfused with 2 l of modified EuroCollins solution at 4°C with a perfusion pressure controlled at 30 cm H2O. Effluent blood and perfusate were vented via the left atrial cannula and discarded. Two litres of topical saline at 4°C were also instilled into the thoracic cavity. Blood collection continued until 2 l had been withdrawn into the perfusion circuit described below. Ventilation with 100% oxygen at a tidal volume of 10 ml/kg continued and the lungs were assessed by perfusion as described below.
For NHBD, the lungs were ventilated with air for 1 min to flush the lungs of excess oxygen prior to clamping of the endotracheal tube to prevent ventilation. Blood withdrawal commenced via the inferior vena caval cannula. The mediastinum was packed with a single sterile swab and the skin over the median sternotomy closed with towel clips. Hypoxic cardiac arrest ensued within 5 min of ceasing ventilation. Blood withdrawal continued until 2 l had been withdrawn; directly into the perfusion circuit reservoir for the 1 and 2 h ischaemic non-heart-beating donor groups. For the 4 h ischaemic non-heart-beating donor group, blood was withdrawn into sterile citrated blood bags to avoid damage from prolonged recirculation within the perfusion circuit.
The lungs were then reinflated with 100% oxygen and the animal left for the required warm ischaemic interval prior to reopening the sternotomy. Ventilation with 100% oxygen at a tidal volume of 10 ml/kg was recommenced and the lungs were assessed by perfusion as described below. Polystan 16 and 21 French gauge perfusion cannulae, respectively were inserted into the main pulmonary artery and left atrial appendage through purse-string sutures. Aortic outflow was occluded by cross-clamping and both lungs perfused with 2 l of modified EuroCollins solution at 4°C with a perfusion pressure controlled at 30 cm of water. Effluent blood and perfusate were vented via the left atrial cannula and discarded. Two litres of topical saline at 4°C were also instilled into the thoracic cavity.
The perfusion circuit, shown in Fig. 2 , comprised an occlusive bypass roller-pump, a cardiopulmonary bypass membrane oxygenator (Dideco Compactflo D703 with 40 µm arterial filter, Sorin Biomedica, W. Sussex, UK) supplied with 95% nitrogen and 5% carbon dioxide gas serving as a deoxygenator, a venous reservoir, a heat exchanger set to 37°C, neutrophil depleting filter (Pall BC1b, Phoenix Cardiovascular, Lancashire, UK), polyethylene tubing (Polystan, Nottingham, UK) and the previously inserted pulmonary arterial and left atrial cannulae. The circuit was primed with 1 l of Hartmann's solution and the haematocrit was standardized to 25%. The pulmonary arterial blood was maintained in the pH range of 7.27.4 by regulation of the carbon dioxide inflow. During isolated ventilation and perfusion assessment, blood gas samples were taken regularly from the pulmonary arterial inflow and left atrial outflow. After perfusion, lung compliance was calculated as the gradient of the range of tidal volumes plotted against peak inflationary pressures during ventilation at that volume. In addition, lung biopsies were taken at various time-points as shown in Table 2 for wet/dry weight ratios; the ratio calculated from the fresh weight divided by the weight after drying for 3 days at 70°C.
|
|
For the 1 h warm ischaemic group (NHBD1) and its control group (Control1), the right hilum was occluded by a large atraumatic clamp and the isolated left lung perfused for 5 min with blood flow controlled at 500 ml/min. These left lungs, having been ventilated and blood-perfused, were then prepared for transplantation.
For the 2 and 4 h warm ischaemic groups (NHBD2 and NHBD4, respectively) and their common control group (Control2,4), the left hilum was occluded by a large atraumatic clamp and the isolated right lung perfused for a longer period of 20 min with the blood flow controlled at 500 ml/min. The left lungs, having only been perfused with modified EuroCollins solution, were then prepared for transplantation.
Left lungs were then transplanted as published in detail previously [18]. In brief, recipients underwent anaesthesia with azaperone (12 mg/kg) and diazemuls (2 mg/kg), jugular and carotid cannulation for monitoring and drug infusion, followed by double cervical tracheostomy for independent ventilation of left and right lungs. Recipients received antibiotic prophylaxis with 750 mg cefuroxime intravenously at induction. Prior to reperfusion, recipients also received methylprednisolone (500 mg) and heparin sulphate (5000 U) intravenously. The mean cold ischaemic time at reperfusion was 7 h and 0 min for Control2,4. For the 2 h warm ischaemia group (NHBD2), the mean cold ischaemic interval was 7 h and 10 min, giving a mean total ischaemic time of 9 h and 10 min.
Once transplantation of the left lung had been performed, four vascular cannulae allowed pressure monitoring of the pulmonary artery and left atrium with blood gas sampling of the pulmonary artery and left lung venous effluent. In addition, blood flow through the left lung was measured independently using an ultrasonic flow-probe. The animals were followed-up under terminal anaesthesia for 12 h prior to euthanasia with sodium pentobarbital (Dolethal, 4 g).
| 3. Results |
|---|
|
|
|---|
The results from assessment using the ventilation and perfusion technique are shown in Figs. 3 and 4 . The 1 h ischaemia group (NHBD1) did not differ significantly from Control1 in terms of oxygenation (ANOVA, P=0.152), or vascular resistance (ANOVA, P=0.157), over the 5-min period of assessment. However, oxygenation differed significantly in the 2 h ischaemia group (NHBD2) compared with Control2,4 (ANOVA, P<0.0005). In addition, the overall oxygenation changed with time, (ANOVA, P<0.0005); oxygenation deteriorated significantly from 8 min onward versus initial values in the NHBD2 group (unpaired t-test with Bonferroni correction, P=0.03 and P=0.003).
|
|
The 4 h ischaemia group (NHBD4) did not behave consistently; four suffered from a high vascular resistance and rapidly developed pulmonary oedema preventing ventilation (shown as NHBD4 poor in Fig. 4). The other two lungs from this group performed well with the assessment technique; their results (shown as NHBD4 good in Fig. 4) mirror those of the Control2,4 group.
Lung compliance data immediately following ventilation and perfusion assessment were as follows: Control2,4, 12.7±1.9 ml/cm H2O; 2 h ischaemia group NHBD2, 12.0±2.6; 4 h ischaemia group NHBD4 lasting the perfusion period (n=3), 13.4±1.1. These did not differ significantly (unpaired t-test).
Transplantation of assessed left lungs from both Control1 and the 1 h ischaemia group NHBD1 revealed uniform poor function, with development of excessively high vascular resistance and poor oxygenation (not shown). Blood flow through the transplanted lungs fell to less than 100 ml/min in all cases by 6 h post-reperfusion; a level not compatible with survival in the clinical setting. For this reason, in subsequent experiments, the right lung was assessed using the technique and the left lung transplanted after cold EuroCollins preservation only, i.e. without ventilation and perfusion assessment.
Fig. 5 shows the left lung transplant results for the Control2,4 and 2 h ischaemia NHBD2 groups. As with the assessment, oxygenation was significantly poorer overall in the NHBD2 group post-transplantation compared with Control2,4 (ANOVA, P=0.001), although there were no significant changes with time (ANOVA, P=0.829). Nevertheless, the levels of pulmonary venous oxygen tension for the NHBD2 group at time-points between 4 and 8 h post-transplantation were still functionally acceptable at a mean of around 30 kPa. Vascular resistances did not differ significantly between the two groups (ANOVA, P=0.986), although both showed highly significant changes overall with time (ANOVA, P=0.001). This was due to the very high initial values, which reduced to more functional levels after 20 min of reperfusion.
|
Fig. 6 shows the relationship between oxygenation during assessment and oxygenation post-transplantation for the 2 h warm ischaemic group NHBD2, the single transplanted 4 h warm ischaemic lung group (NHBD4) and the Control2,4 group.
|
|
| 4. Discussion |
|---|
|
|
|---|
The performance of lungs subjected to EuroCollins and neutrophil-depleted blood perfusion limited to 5 min was very poor, even with lungs from heart-beating control donors. However, as mentioned above, the data shown from the 2 h warm ischaemia group transplanted after EuroCollins perfusion alone showed adequate post-transplantation performance. Based on this finding, the authors suggest that either the combination of EuroCollins and blood perfusion, or the blood perfusion alone, is detrimental to lungs retrieved from the non-heart-beating donor. Whether this was due to the EuroCollins or the blood perfusion alone remains to be elucidated. For this reason, we would recommend the assessment of one lung and transplantation of the contralateral organ assuming the absence of focal disease, as both organs will have suffered an identical degree of warm ischaemic injury.
Clinical and experimental work in lung transplantation has shown us the importance of careful assessment of the potential donor lung. Many physiological parameters such as airway pressure, pulmonary vascular resistance (by inference from right ventricular performance and measured left atrial pressure during retrieval) affect the overall decision on whether to accept or reject the lung for transplantation and the oxygenating index is probably the most sensitive of these tests [16]. These factors can be assessed with our proposed assessment technique.
Two areas of this research are open to debate, concerning firstly the arrest model used for non-heart-beating lung donation and secondly the transplantation of only one of the contralateral lungs perfused in the 4 h ischaemic group NHBD4.
The arrest model was designed to be similar to the likely clinical scenarios of the non-heart-beating donor. A period of apnoea induced by cessation of ventilation would occur following major intracerebral haemorrhage or severe head injury. The subsequent cardiac arrest may be preceded by a short period of strain on the pulmonary circulation, in turn resembling the clinical primary cardiac arrest scenario of agonal left ventricular failure. These may cause some pulmonary ischaemia or oedema, affecting the pulmonary function in a hopefully realistic manner. This was felt necessary to test the feasibility of the assessment technique, rather than to develop a model with absolutely preserved lung function offering no degree of injury to assess.
The use of intravenous potassium chloride solution to induce cardiac arrest by rapid asystole for the model was considered but rejected for two reasons. Firstly, such a rapid cardiac arrest was felt to be a little too benign in terms of the effects on the lung, and secondly, the raised potassium concentrations in the collected blood would have caused problems with vasoconstriction at reperfusion during assessment.
Ventilation with ten breaths of 100% oxygen 10 min after cardiac arrest was intended to resemble a short resuscitation, and inflation with oxygen during the warm ischaemic interval has previously been shown to optimize function of the non-heart-beating donor lung. The time-period of 10 min was chosen in view of the Maastricht guidelines for a clear 10-min interval between cessation of resuscitation and any procedures taken solely for organ preservation [20].
The use of blood from the same donor as the lungs would reduce the pulmonary blood volume and may also affect the lung function; there are reports of pulmonary inflammation associated with haemorrhage in animal models [21]. This decision was taken to avoid the potential confounding problems of a transfusion reaction with the use of a third animal as a blood donor, and as a secondary factor, to reduce the numbers of animals used in the series of experiments. However, the same blood collection technique was used for both of the control groups so this is unlikely to confound the effect of non-heart-beating donation.
To take the second area of weakness, only one of the six lungs from the group with 4 h warm ischaemia (NHBD4) was transplanted. This lung had performed as well as controls (Control2,4) during assessment and functioned satisfactorily post-transplantation. It is accepted that poor results post-transplantation for those four lungs with very poor performance under initial assessment would have increased the strength of the findings of the work. However, these four lungs all developed gross haemorrhagic pulmonary oedema during perfusion, unable to be ventilated or perfused without excessive airway pressure or perfusion pressures. The authors considered that these badly damaged lungs would not have the potential to function on reperfusion after the further insult of a cold ischaemic period and transplantation. In addition, the strict Home Office licensing rules in the UK prevented us from transplanting any organ with clearly poor function to prevent undue suffering to the terminally anaesthetized recipient.
Although the ventilation and perfusion circuit technique has been used for many years in lung research, predicting the function of the paired lung post-transplantation is a novel use. Parallels clearly exist with the perfusion assessment of the non-heart-beating donor kidney. We think that the technique, as a predictor of function, may be vitally important for a clinical non-heart-beating donor transplant programme.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
H. Inokawa, M. Sevala, W. K. Funkhouser, and T. M. Egan Ex-vivo perfusion and ventilation of rat lungs from non-heart-beating donors before transplant. Ann. Thorac. Surg., October 1, 2006; 82(4): 1219 - 1225. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. M. Egan, J. A. Haithcock, W. A. Nicotra, G. Koukoulis, H. Inokawa, M. Sevala, P. L. Molina, W. K. Funkhouser, and B. J. Mattice Ex Vivo Evaluation of Human Lungs for Transplant Suitability Ann. Thorac. Surg., April 1, 2006; 81(4): 1205 - 1213. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. S. Wilkes, T. M. Egan, and H. Y. Reynolds Lung Transplantation: Opportunities for Research and Clinical Advancement Am. J. Respir. Crit. Care Med., October 15, 2005; 172(8): 944 - 955. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. R Rega, E. J Vandezande, N. C Jannis, G. M Verleden, T. E Lerut, and D. E. Van Raemdonck The role of leukocyte depletion in ex vivo evaluation of pulmonary grafts from (non-)heart-beating donors Perfusion, January 1, 2003; 18(1_suppl): 13 - 21. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |