|
|
||||||||
a Department of Surgery, Washington University, St. Louis, Missouri
b Department of Medicine, Washington University, St. Louis, Missouri
Accepted for publication June 4, 2009.
* Address correspondence to Dr Meyers, Division of Cardiothoracic Surgery, Barnes-Jewish Hospital, Queeny Tower, 3108, One Barnes-Jewish Hospital Plaza, St. Louis, MO 63110 (Email: meyersb{at}wudosis.wustl.edu).
Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
| Abstract |
|---|
|
|
|---|
Methods: We performed a retrospective review of DCD experience at a single lung transplant program using a prospective database.
Results: Between January 2003 and April 2008, 293 lung transplantations were performed, including 11 bilateral transplantations (3.7%) using DCD lungs. Similar criteria were used to assess donor quality. The hospital mortality for DCD recipients was 2 of 11 (18%) and overall mortality was 4 of 11 (36%) by 18 months of follow-up. Seven DCD patients (64%) are alive with a median follow-up of 32 months. The DCD group was comparable to the control group in age and ischemic times. The 4 deaths, when compared with 7 DCD survivors, had longer ischemic time (293 minutes versus 232 minutes) and a higher incidence of nonlocal donors (3 of 4 versus 1 of 7).
Conclusions: At our center, early outcomes after DCD lung transplantations are somewhat inferior to those of series from other centers but approach national averages for conventional lung transplantation. Thus, DCD lung transplantation has the potential to increase the donor pool but must be offered cautiously.
| Introduction |
|---|
|
|
|---|
It has been shown in experiments that lungs and their vascular function can be safely preserved for as long as 24 hours [7, 8]. The gas-exchange system of the lungs can tolerate 1 hour of warm ischemia after circulatory arrest while still maintaining its functional integrity [2, 9], and the pulmonary artery can withstand warm ischemia for 3 hours after death without impairment of endothelium-dependent relaxation or vascular smooth muscle function [10]. Additionally, mere topical cooling of nonventilated lungs gives excellent preservation for 12 to 24 hours [11]. De Antonio and colleagues [12] published the initial clinical report of 17 patients when their group began offering lungs obtained from uncontrolled nonheart-beating donors to patients in 2002 after a preclinical study had shown the functional and histologic viability of these grafts [13].
The lung transplant community has been slow to adopt donation after cardiac death (DCD), primarily because of the concern that increased warm ischemia might produce organ injury and graft dysfunction. This concern has resulted in only a handful of publications describing the clinical results of lung transplantation using DCD [12, 14–18], and no single center has a large experience. A recent paper summarized the US DCD lung transplantation experience from the UNOS database [19]. Looking at our own experience, our impression was that our results differed and were not entirely captured by the United Network for Organ Sharing database, leading to the present publication.
| Material and Methods |
|---|
|
|
|---|
At our program, the criteria for evaluation of DCD donors were similar to those for conventional donors, and are outlined in Table 1. Donor lungs were considered "local" when they originated within the St. Louis metropolitan area and could be transported by ground transportation to our hospital. Other lungs from outside our region and for which air travel was required were classified as "distant" donor lungs. We considered both local and distant donors for DCD donations if the expected flight time was less than 60 minutes. Another classification system labels the lungs as either "ideal" or "marginal" based on important clinical characteristics of the donor. The key elements of this classification are reported in Table 2. Marginal donors, as characterized by this system, were not considered suitable for DCD lung procurement. There was a conscious attempt to use the DCD lungs for recipients considered to be at baseline risk for perioperative complications. As a result, recipients with emphysema, cystic fibrosis, and bronchiectasis were considered for the procedure, whereas recipients with pulmonary hypertension and pulmonary fibrosis were not. Patients with previous chest surgery or pleurodesis were excluded owing to unpredictable explant procedures. The entire lung transplant team, including surgeons, transplant pulmonologists, and transplant coordinators, was briefed at the initiation of the program and periodically thereafter.
|
|
Postoperative Management
The postoperative management was identical to that provided to a standard lung transplant patient, with emphasis on maintaining a low central venous pressure (less than 7 cm H2O) and weaning from mechanical ventilation as permitted. The patients were generally extubated within 24 hours if PaO2/FiO2 was more than 200. Primary graft dysfunction was defined and managed per standard guidelines [22].
Immunosuppression therapy, surveillance, antibiotic prophylaxis, and diagnosis of bronchiolitis obliterans were no different than that used for conventional transplant recipients.
Statistics
Retrospective review of the DCD experience at a single lung transplant program using a prospective database between January 2003 and April 2008 was performed. Approval for the study was obtained from the Institutional Review Board, and need for individual consent was waived owing to the minimal risk of this retrospective review. Data are expressed as means unless otherwise stated. Descriptive statistics are presented because small numbers preclude meaningful statistical comparisons between the groups.
| Results |
|---|
|
|
|---|
|
|
|
| Comment |
|---|
|
|
|---|
Our DCD perioperative mortality (2 of 11, 18%) is higher than our perioperative mortality from conventional donors (11 of 282, 3.9%) and that seen in outcomes from large databases [19]. Our first perioperative death had a distant donor and relatively long cold ischemic time (335 minutes). She had grade 3 PGD, required prolonged ventilator support, and died of fungal sepsis. Our second perioperative mortality had a local donor and relatively short ischemic time (207 minutes), but severe PGD developed requiring extracorporeal membrane oxygenation (ECMO) support. The patient was weaned from ECMO but died of abdominal sepsis at 27 days. In our series, the 4 intermediate-term deaths had longer ischemic times (293 versus 232 minutes) and were more likely to have a distant donor (3 of 4 versus 1 of 7). The absolute number of patients in our series is small, and any differences may be due to chance. Snell and associates [17] noted a mean overall ischemic time of 443 minutes, and Mason and associates [20] reported an average maximum lung ischemic time of 371 minutes in the DCD cohorts reported by these authors. Both series describe favorable outcomes, and thus we cannot place too much emphasis on the differences in ischemic times in our cohort. The mean ischemic time in our report is the average of the two ischemic times in double lung transplantation, hence apparently lower than the numbers in these two reports.
We have avoided the use of DCD lungs for patients with pulmonary hypertension and pulmonary fibrosis and for patients with prior pleurodesis to avoid long and unpredictable explant procedures. This policy favors patients with emphysema, cystic fibrosis, and bronchiectasis who may have lower lung allocation scores and may not get transplanted otherwise. Also, this approach selects a group of patients who are at a baseline risk for complications, avoiding the higher risk population of patients with pulmonary hypertension and pulmonary fibrosis.
The 4 deaths were of patients who suffered from significant perioperative morbidity, with 2 of them requiring ECMO support for severe PGD. All 4 also had renal failure. Of the 7 survivors, only 1 had renal failure that did not require dialysis. There was 1 major airway complication of the 22 anastomoses at risk. The patient required a sleeve right upper lobectomy and has done well since. Our airway complication rate of less than 5% is similar or favorable to that of published data [23]. The incidence of grade 3 PGD is 36% in the DCD group and 18% in our general transplant cohort (Table 4). This difference may be due to chance as the numbers in the DCD group are small. Snell and colleagues [17] describe an incidence of 13% PGD grade 3 in their DCD population whereas deAntonio and colleagues [12] noted an incidence of 29%. It is suspected that warm ischemia may contribute to the incidence of PGD in the DCD population, but the precise nature of the impact is not clear with the limited clinical data available.
Snell and colleagues [17] have demonstrated excellent short-term results from their series of 8 DCD Maastricht 3 lung transplants, with no perioperative mortality and 1 patient requiring perioperative use of ECMO for grade 3 PGD. Another recent paper [19] summarizes the North American experience from the UNOS database looking at 36 patients with an excellent survival of 87% at 24 months in DCD lung transplantation.
Reviewing our own patients and comparing them to the published UNOS database results, which should be inclusive of data from all US transplant centers [19], 2 of our intermediate-term deaths (51 days and 13 months) were not captured by the UNOS database. Our own 1-year survival of 81% and 2.5-year survival of 63% are closer to the outcomes seen for conventional bilateral lung transplantation in the International Society for Heart and Lung Transplantation registry (83% 1-year survival, 67% 3-year survival) [24].
In the context of DCD, there is no clear consensus on the definition of warm ischemia. Oto and colleagues [14] defined it as the interval between cessation of cardiac output and the time to antegrade pulmonary artery flush. In addition, there is a second period of warm ischemic time during implantation: the time from the removal of the lung from the cold preservative to the time of restoration of blood flow to the graft lung. It is generally recommended that the initial warm ischemia not exceed 60 minutes [17]. Although we did not formally record warm ischemic time in our database, every procurement involved withdrawal of support in the operating room, and there were no instances of prolonged cardiac activity after extubation. It is likely that our initial warm ischemic times were in the 20- to 45-minute range.
Our series has obvious limitations. It is a single-center, retrospective review of a relatively small patient population. Also, we do not have recorded data for warm ischemic times. Still, because our data add to a growing body of literature about DCD lung transplantation, which can increase the donor pool, we believed it would be of interest to the lung transplant community.
In conclusion, our patient outcomes for DCD lung transplantation show acceptable survival in the intermediate term. Hence, DCD lung transplantation has been shown to be an additive influence on the number of patients undergoing lung transplantation [17] and may continue to be cautiously offered as part of a lung transplant program.
| Discussion |
|---|
|
|
|---|
DR PURI: Sixty minutes. We use the standard DCD evaluation tool to predict the probability of this.
DR FORCE: Did you go back and look at that time in any of the patients who did poorly to see if it was closer to the 60 or to the 20 or 15 minutes?
DR PURI: Of the 11 patients, we have warm ischemic times recorded in 6. In all of them, the ischemic times are between 15 and 35 minutes. For the remaining 5 patients—since it's a very small patient population, everyone remembers these patients quite well—in none of them was prolonged cardiac activity observed after extubation. So our warm ischemic times were not a problem.
DR FORCE: Just two more quick questions. Number one, when I was at Barnes, the method of procuring for DCD donors was just to wait and do it in the standard fashion once the heart had stopped. Other authors have proposed putting in chest tubes as soon as the patient is extubated and irrigating with cold saline. Have you changed your techniques to try to reduce the incidence the primary graft dysfunction?
DR PURI: No. It is still similar to what you had observed when you were there. It's fairly similar to conventional lung transplantation procurement technique. And since the warm ischemic times are low as we see, I don't think it would make a difference.
I believe the experience that you're referring to is D'Antonio's paper from the 2007 Journal of Heart and Lung Transplantation in Maastricht I and II donors in which they used this protocol and irrigated the thoracic cavity with the cold saline. Also some other programs are implementing a Maastricht III donor program in which they withdraw support in the intensive care unit. That would possibly increase warm ischemic times.
DR FORCE: My final question is for the chairs as well as you. I read a lot of abstracts and papers on DCD donors. I do not get a lot of offers for DCD donors. These patients are usually in the hospital for a long time. They develop multiple complications, pneumonias. And my thought has always been whenever I read an abstract, is this really going to increase the lung donor population that much? You've mentioned other strategies that I think are much more likely to. I want to get your thoughts. I mean, I'm always happy to use a DCD donor. I haven't done one in 5 years just because I haven't gotten a suitable donor.
DR JOSHUA R. SONETT (New York, NY): I think this paper is excellent because it introduces a word of caution in regard to a new technique that is very exciting and especially being used more commonly. At Columbia, we've just done 2 DCD donors mostly because in this early phase, as with St Louis, We have been very picky in regard to the donors. We have limited it to 20- to 30-year-olds, limited ischemic time, with less travel. But I think as we become more comfortable with the technique we may be able to add 6% or 10% to the donor pool, and it will really help. We may just have to be careful not to push the envelope too hard as this paper shows.
Now, let me ask you. What are your criteria now? Have you changed at all because you looked at it? Which ones would you take versus not take for DCD donors?
DR PURI: Before I answer that, do you mind if I add two pieces of data to Dr Force's question. We have received about 63 offers in this period of time, of which we have accepted 11 DCD lungs. And Dr Snell from Australia has shown that it has increased their annual transplantation numbers using DCD by about between 10% and 15%.
Now, for criteria, we haven't really changed our criteria because they were fairly strict to start with. I do believe that of the 2 the patients who died with primary graft dysfunctions, 1 of them was likely related to chance. The other had significantly prolonged ischemic times. And others have shown that in their limited DCD experience, these patients do quite well. I'm privy to some data from the University of Wisconsin, which are going to be published or presented to the AATS by Dr DeOliveira, and his numbers are better than ours. So I do think that experienced lung transplant programs should proceed cautiously with this, but again, maintaining very tight controls over the quality of the donor organ and taking it from patients that are safe.
DR MICHAEL S. MULLIGAN (Seattle, WA): Liver and kidney programs are backing away from this. You're going to see the DCD numbers go down for 2009. The cautionary notes about biliary strictures and problems with timing of withdrawal are big. Your protocol appears to be in line with what we arrived at the Consensus Conference in Chicago 3 years ago, but I think I would add some practical notion to what Seth is asking.
I would only do local DCD donors. I think you can take that away from this. And I think you need to put your most experienced team in the field to do this. If you put a junior fellow out in the field and ask him to flounder around with an abdominal surgeon who wants to vent in the chest, as is the routine with DCD abdominal donation, you're going to get into problems.
So it needs to be very tight, and you should have a very low threshold for walking away. Having an attending in the field will assist that. Having a resident who is reluctant to make that decision and stop the transplant process at his discretion may lead you to carry that momentum forward when you would rather arrest it.
DR BRYAN FITCH MEYERS (St. Louis, MO): Those are good points, and you also might even consider the notion of additional consent when you use a DCD donor. We have to do additional consents now for donors who are considered socially high risk donors. This DCD situation is another example of when the donor is distinct from the rest of the donor pool and might require extra caution and explanation to the potential recipient.
DR SONETT: I think it's a requirement, actually, for CMS. You need to separately consent for anything out of the norm, and certainly DCD is out of the norm.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. R. Manara, P. G. Murphy, and G. O'Callaghan Donation after circulatory death Br. J. Anaesth., January 1, 2012; 108(suppl_1): i108 - i121. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Van De Wauwer, E. A. M. Verschuuren, W. van der Bij, G. D. Nossent, and M. E. Erasmus The use of non-heart-beating lung donors category III can increase the donor pool Eur J Cardiothorac Surg, June 1, 2011; 39(6): e175 - e180. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Kreisel, A. S. Krupnick, V. Puri, T. J. Guthrie, E. P. Trulock, B. F. Meyers, and G. A. Patterson Short- and long-term outcomes of 1000 adult lung transplant recipients at a single center J. Thorac. Cardiovasc. Surg., January 1, 2011; 141(1): 215 - 222. [Abstract] [Full Text] [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 |