|
|
||||||||
Ann Thorac Surg 2006;82:1835-1841
© 2006 The Society of Thoracic Surgeons
a Department of Surgery, University of Virginia, Charlottesville, Virginia
b Department of Public Health Sciences, University of Virginia, Charlottesville, Virginia
c Department of Internal Medicine, University of Virginia, Charlottesville, Virginia
Accepted for publication May 11, 2006.
* Address correspondence to Dr Jones, Department of Surgery, University of Virginia, PO Box 800679, Charlottesville, VA 22908-0679 (Email: djones{at}virginia.edu).
Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30Feb 2, 2006.
| Abstract |
|---|
|
|
|---|
60 years could be performed with acceptable outcomes. METHODS: We identified 182 consecutive LTX recipients from 1995 to 2005. Outcomes were analyzed and survival compared with results in recipients aged <60, as well as with United Network for Organ Sharing (UNOS) registry outcomes for the same age and study period. Actuarial survivals were calculated by the Kaplan-Meier method.
RESULTS: During the study period, 29% (52/182) of LTX recipients were
60 years old (range, 60 to 69 years). Median follow-up was 2.9 years (range, 0 to 10 years). All patients but one received a single lung. Indications included chronic obstructive pulmonary disease in 63% (33/52), idiopathic pulmonary fibrosis in 27% (14/52), and other in 10% (5/52). In-hospital mortality was 12% (6/52) for those aged
60 compared with 7% (9/130) for those aged <60 (p = NS). Complications included reoperation in 10% (5/52), requirement for extracorporeal membrane oxygenation in 6% (3/52), renal failure in 12% (6/52), and stroke in 4% (2/52). Actuarial survivals at 30 days, and 1, 3, and 5 years were 90% (82, 98), 86% (76, 96), 71% (56, 85), and 55% (37, 73), respectively. No significant difference in survival was observed between age cohorts for our institutional data by Kaplan-Meier analysis (p = 0.34) or by Cox proportional hazard model (p = 0.15). A significant survival advantage was noted for our institution compared with UNOS for this cohort (p = 0.018).
CONCLUSIONS: In carefully selected recipients
60 years of age, LTX offers acceptable outcomes and survival.
| Introduction |
|---|
|
|
|---|
Several reports examine factors predictive of death after LTX, one of which is advanced recipient age [57]. In the early era of organ transplantation, LTX in older patients was avoided owing to their anticipated high risk from both the procedure and lifelong immunosuppression. By the end of the 1980s, multiple publications were supporting kidney, liver, and heart transplantation in recipients >50 years. In 1993, Snell and colleagues [8] published the first review of the experience in LTX in patients aged >50 years, with only 5 of 103 patients >60 years. They found no difference in survival or functional outcome between groups.
Advanced recipient age continues to be used as an exclusion criterion for LTX [9, 10]. A 1998 consensus conference of thoracic and transplant societies produced guidelines for the selection of recipients for LTX that include a recommended recipient age limit of 65 years for single-lung transplantation (SLT) and 60 years for bilateral lung transplantation (BLT) [9]. These guidelines are not absolute, and actual practice varies between centers. In a survey of 50 active LTX centers published in 2004 [10], selection criteria were consistent overall, but center-specific practices for age exclusion varied. For BLT, 29% of programs considered age >55 years to be a relative contraindication, whereas only 24% and 4% of centers do not consider age to be a contraindication to BLT in patients age >60 and >65 years, respectively. In contrast, less than 20% of centers do not consider age >65 years to be a contraindication to SLT.
The question of appropriate recipient age for LTX is pressing, given the changing age demographics in North America and Europe. In 2000, more than 45 million people in the United States (US) were
60 years of age, representing more than 16% of the population. As our population ages, this proportion will continue to grow. It is projected that by 2030, more 25% of the US population will be aged
60, with almost 20% of those being
65 (US Census Bureau). If a strict age limit of 60 years were used for LTX, a quarter of our population would be categorically denied this therapeutic option.
In keeping with this aging patient population, national and international registry data demonstrate that the age distribution of adult LTX recipients has also shifted significantly. The proportion of recipients aged
60 years has expanded noticeably, and this age group now comprises
20% of adult recipients since 1997 [7].
Review of the literature demonstrates no previously published study that has focused specifically on outcomes for LTX recipients
60 years of age. Over a 10-year period, our center performed 182 consecutive LTX. A growing percentage of recipients were
60 years old, with a cumulative 29% of our recipients aged
60 at the time of LTX (Fig 1). We hypothesized that the outcomes for our LTX recipients aged
60 are not significantly worse than those for recipients <60 years old, and furthermore, that advanced age should not be considered a contraindication to LTX in otherwise appropriate recipients.
|
| Material and Methods |
|---|
|
|
|---|
Standard LTX procedures were performed. Cardiopulmonary bypass (CPB) was used only as required. All recipients received induction therapy. Earlier recipients received antithymocyte globulin (Pharmacia Upjohn, Kalamazoo, MI), and after 2001, recipients received Daclizumab (Roche, Ltd., Mississauga, Canada). In the perioperative period, recipients were managed primarily by the surgeons, with the aid of other members of the multidisciplinary LTX team. Recipients were divided into two groups; those aged <60 years, and those aged
60 years on the date of LTX. Pretransplant characteristics, operative variables, in-hospital complications, and overall survival were examined.
The United Network for Organ Sharing (UNOS) maintains records on organ donation and transplant events occurring in the United States since 1986. UNOS provided the Standard Transplant Analysis and Research Dataset for LTX during our period of interest. These data were used where appropriate to provide a national perspective and as a comparison with our institutional experience.
Statistical Analyses
All statistical analyses were performed using SAS 9.0 software (SAS, Cary, NC). To test whether associations existed between groups and outcomes for categoric variables, the
2 test or the Fisher exact test were used where appropriate. The two-sample t test was used to compare group means, and the nonparametric Wilcoxon rank sum test was used to compare the medians between groups for continuous variables. For the survival analyses, the Kaplan-Meier survival estimate was used to obtain the survival curves, and standard confidence intervals were calculated at specific time points. The log-rank test was used to compare the survival functions between groups. A Cox proportional hazards model was fit to investigate the effect of age on survival, with adjustment for multiple covariates. All statistical analyses used a significance level of 5%, and the validation of the assumptions underlying the statistical methods was checked with standard statistical diagnostic tools.
| Results |
|---|
|
|
|---|
60 years old, and 9% (16/182) were
65 years old. The age groups had similar characteristics with the exceptions that the age
60 group had a larger smoking burden and more idiopathic pulmonary fibrosis (IPF) and chronic obstructive pulmonary disease (COPD). The characteristics of the recipients aged
60 at our institution were similar to the national group aged
60, with the exception that our institution performed a greater proportion of LTX for sarcoidosis. The median age at listing of the patients in our age
60 recipient group was 61 years, with 73% (38/52)
60 years old when they were listed. Of the recipients aged
65, 50% (8/16) were
65 years at the time they were listed.
|
60 received a SLT with one exception. This patient received a BLT for COPD with pulmonary hypertension. Our institution has preferentially performed SLT rather than BLT for recipients aged
60 at a significantly higher rate than nationally.
Institutionally, there was no difference between age groups for length of stay (LOS), time of ventilator support, or intensive care unit (ICU) stay; nationally, the age
60 group had a 1 day shorter LOS than the age <60 group (p < 0.0001). For all complications examined, the two age groups were not significantly different, with the exception of more frequent dysrhythmias in the age
60 group. In-hospital mortality was 12% (6/52). Causes of mortality were sepsis (n = 1), cerebrovascular accident (n = 1), primary graft dysfunction (n = 3), and myocardial infarction (n = 1). Morbidity included 17 dysrhythmias (11 survivors), nine pneumonias (7 survivors), two cerebrovascular accidents (1 survivor), six episodes of acute renal failure (2 survivors), four episodes of deep venous thrombosis with two pulmonary emboli (all survivors), and three requiring extracorporeal membrane oxygenation (ECMO) support (no survivors).
Five recipients aged
60 required reoperation. Indications included immediate reexploration for a contralateral (endotracheal tube related) mainstem bronchial tear recognized on intraoperative bronchoscopy, contralateral lung volume reduction on postoperative day 8, a negative reexploration for questionable left atrial anastomotic thrombus seen on echocardiography, reexploration in the ICU for open cardiac massage after cardiovascular collapse from a contralateral tension pneumothorax, and reexploration for bleeding. The only survivor after reoperation in this age cohort was the patient with the contralateral bronchial injury.
Figure 2
demonstrates that our institutional outcomes compare favorably with national results as recorded in the UNOS registry for both age groups. For recipients aged
60, we found a significant survival advantage for LTX performed at our institution (Table 2). We also looked separately at our 16 recipients who were
65 years old at LTX (Table 2). Indications for LTX in this group were COPD (n = 7), IPF (n = 7), sarcoidosis (n = 1), and other (n = 1). Median follow-up was 1.7 years (range, 1 to 6.6 years). A significant institutional survival advantage is seen compared with UNOS data. Although this is a limited sample size, there was better than 90% 5-year actuarial survival, with 5 patients having reached at least 5 years of follow-up. One death occurred in this group in 43 patient follow-up years.
|
|
60 than those aged <60 (Fig 2), which is statistically significant (p < 0.001). Analysis of our institutional data (Fig 2 and Table 3) reveals no significant difference in survival between the two age groups over the entirety of the study period by the Kaplan-Meier method based on the log-rank test. We chose an age cutoff of 60 years as a practical division between groups based on current practice. We also examined the survival effect of age as a continuous variable, however. A Cox proportional hazards model was fit to the University of Virginia data with covariates of age, body mass index (BMI), gender, diagnosis, and pack-year smoking history. This analysis revealed a hazard ratio of 1.029 per year for the effect of age, which was not statistically significant (p = 0.15).
|
60 had a trend towards poor survival specifically to 30 days, we performed Kaplan-Meier calculations conditional on survival past 30 days (Fig 3
and Table 4). For those recipients who survived to 30 days, the age <60 and the age
60 groups had very similar survival. There was minimal absolute difference and, again, no statistically significant difference between the groups (p = 0.61).
|
|
60. Therefore, we examined survival specific to these indications (Table 3). The survivals were statistically equivalent between the age groups after LTX for the diagnosis of IPF in this limited group. COPD accounted for two thirds of the LTX performed in our recipients aged
60. The survival differences between the age groups are significant for COPD. However, after an initial 9% mortality by 30 days in the age
60 group, the subsequent survival was similar between the age groups based on a 30-day conditional Kaplan-Meier survival analysis for this group. With the exception of one death in the age
60 group at 31 days after LTX, no further deaths occurred in either group until 1.5 years after LTX, and the difference in survival between age groups was not significant (p = 0.12).
During the second half of our study period, 401 patients were referred to our program for consideration for LTX; 70% (279) were <60 years, and 30% (122) were
60 years at the time of referral. The age range of all referrals was 5 to 81 years. Of those referred, 51% (208) were considered appropriate candidates and were listed for LTX. The percentage of referrals who were placed on the waiting list was similar between the age groups; 53% (148/279) of those <60 years and 48% (58/122) of those
60 years (p = 0.33) were eventually listed for LTX. There was no difference between the two age groups for the incidence of death while they were being evaluated or the incidence death while on the waiting list.
With regards to the reasons that patients were not listed for LTX, there was no difference in the rates of obesity, psychosocial problems, infections, or current smoking as contraindications when the <60-year-old and
60-year-old groups were compared. There was a trend towards higher rates of malignancy and cardiovascular conditions as contraindications in the
60-year-old group, but these did not reach significance (p = 0.19 and p = 0.41, respectively). The
60-year-old age group was more likely not to be listed because the patient declined the procedure than in the <60-year-old group (8% versus 19%, p = 0.04).
| Comment |
|---|
|
|
|---|
60 compared with those aged <60 and, remarkably, excellent outcomes even for those recipients aged
65 years. This study focused specifically on LTX in this growing patient population and validates the ongoing practice of performing SLT in recipients aged well beyond 60 years. Based on these results, we believe that the rapidly growing
60-year-old segment of our population should not be denied LTX because of age alone.
We perform SLT rather then BLT almost exclusively in
60-year-old recipients, but there is practice variation between centers in the selection of SLT versus BLT. Nationally, between 1993 and 2003, BLT more than doubled for both COPD (16% to 38%) and IPF (17% to 38%) for all LTX recipients. This may have been motivated by better collective survival results (particularly for COPD) or by institutional preferences and practices [7]. Although BLT may afford greater long-term ability to tolerate pulmonary decompensation, it potentially is associated with a greater morbidity and mortality, particularly in the older patient.
In 2001, Meyer [11] published a report examining the influence of age on outcomes of SLT versus BLT in COPD based on the North American experience. In that series, 21% of the recipients were >60 years old. Recipients aged
60 years old had no difference in early survival with SLT versus BLT, but there was significantly improved longer-term survival with BLT. By contrast, recipients aged >60 years had much better 30-day survival with SLT than with BLT (93% versus 78%), which remained true until at least 3 years post-LTX. Based on this data, our current institutional policy is to perform BLT for those patients with COPD who are <60 years old and to perform SLT for most of the older patients.
The LTX population carries significant comorbidities, and despite good overall outcomes, it is not surprising that significant complications are observed across age groups. We observed no difference in morbidity after LTX between the <60-year-old and
60-year-old groups, with the exception of dysrhythmia occurring 2.5 times more frequently in the
60-year-old recipients. Few reports are available on dysrhythmia after LTX, particularly for adults. A recent retrospective review of 200 LTX recipients with a mean age of 50 years (range, 19 to 66 years) reported that postoperative atrial fibrillation and flutter together occurred in 39%, with the highest prevalence in those receiving a transplant for IPF (56%) and COPD (42%) [12]. Recipient age
50 was a significant predictor, and those in whom dysrhythmia developed had longer LOS, greater in-hospital mortality, and lower long-term survival. There are no studies to date about perioperative dysrhythmia prevention in LTX, but older patients represent a potential target for investigation in this area.
For those receiving LTX for COPD, recipients aged <60 had a slight survival advantage. This is attributable to higher, though not significant, 30-day mortality in the age
60 group. In the UNOS data, 30-day mortality after LTX was 7%, and in published series, perioperative mortality is as high as 14%, with the major reported causes of early death being early graft failure, cardiac causes, and infection [7, 13, 14]. Our series had 7% (12/182) 30-day mortality for all ages and 10% (5/52) for recipients aged
60. The age
60 group had another death that occurred on postoperative day 31. Including this event, all six early mortalities in the age
60 group occurred during the index hospitalization. The 30-day conditional survival distributions suggest that if potential LTX recipients aged
60 are appropriately selected, good long-term outcomes can be expected. It also highlights that even with careful selection, these older patients do not tolerate major complications well and the need for reoperation is clearly a negative prognostic indicator.
Several predictive risk factors for 30-day mortality and prolonged ICU stay have been identified after LTX [13]. These include use of CPB and a BMI > 25 kg/m2, among others. For our age
60 recipients, five (83%) of six of those who died in-hospital had a BMI > 25 compared with 18 (39%) of 46 of those who survived. The only patient with BMI < 25 who died required CPB. Therefore, all of our in-hospital mortalities in the age
60 group met criteria for high risk. Our experience, combined with that of others, suggests that patients with an age
60 years and a BMI > 25 may not be appropriate candidates for LTX.
The 2005 report of the International Society of Heart and Lung Transplantation identifies multiple pre-LTX variables, other than recipient age, that are predictive of 1- and 5-year mortality after LTX [7]. Relevant to this discussion, recipient factors included ventilator or inotrope dependence, prior sternotomy or thoracotomy, repeat LTX, BMI > 25, elevated bilirubin or creatinine levels, and elevated pulmonary arterial pressure or resistance. There is also a center volume effect with increased mortality at centers performing few LTX procedures. These factors were not stratified for age and should be considered in combination when evaluating any recipient for a potential LTX.
This report is subject to limitations inherent with any single-center retrospective study. The sample size is limited, and some differences between groups may be statistically significant with a larger sample size. This could lead to a type II error. Differences in patient population, surgical technique, and postoperative and outpatient management might lead to different outcomes across different centers, although most LTX centers use similar patient management approaches. Any study of LTX recipients is subject to the selection bias inherent to the screening and evaluation process. Although patient age is not an independent exclusion criterion at our center, older patients undergo the same pretransplant evaluation, and therefore, many are excluded. Of note, our institution has no significant difference in exclusion rate for those aged <60 years compared with those aged
60 years.
In conclusion, we believe that age alone should not be considered a contraindication to LTX in an otherwise appropriate candidate. This study demonstrates that LTX can be performed with very acceptable immediate and long-term results in appropriately selected patients aged
60 years. Furthermore, although the experience is small, recipient age
65 should not be an absolute contraindication to LTX. The risk factors for death we have described should be included in considerations for LTX in these patients, as they should for potential recipients of any age. Further study of how individual risk factors vary or not with age will allow improved risk prediction and selection of LTX recipients.
| Discussion |
|---|
|
|
|---|
The other question would be, if these other databases find that recipient age is a predictor of outcome and you haven't, then you have to pick between one or more of three different conclusions. One, that you have a different selection process for transplantation: you pick healthier older patients or you pick healthier donor lungs for your older patients. Number two, you do the transplant operation differently or have some different method of perioperative care that you didn't let us hear about during the presentation. Or, number three, the one that you had touched on and the one that I'm leaning towards, is that you lack the statistical power to show a difference between a 7-percent and a 12-percent mortality. I'll just leave those questions with you. Congratulations on your presentation.
DR SMITH: Thank you. There have indeed been large series that have examined risk of mortality following lung transplantation using age as a continuous variable, including the registry report of the ISHLT data from 2005. This showed increasing recipient age to be associated with increased mortality at 1 and 5 years following transplant. We chose to use specific age cutoffs for two reasons. First, our sample size limited our ability to do otherwise. Second, we know that lung transplant centers across the country are using specific age limits to determine eligibility for lung transplantation, as was demonstrated in Levine's survey published in Chest 3 years ago. We wanted to evaluate whether this existing clinical practice would correlate with outcomes in our experience.
Your question of why our outcomes are better than other series is certainly an important one. It would be valuable if we could report that we are doing something specifically different and therefore offer guidelines for recipient selection and patient management. For patient selection, when we looked at our patient variables, there weren't many significant differences between our patients and the national data set. Likely the most important difference in management between our series and others is that we performed single lung transplantation almost exclusively for those patients age 60 years and older. There was only one exception out of the 52 patients.
In 2001, Meyer and colleagues published a paper that specifically examined the outcomes of single versus bilateral lung transplantation based on recipient age. That report looked specifically at the US experience in patients with COPD. They found that for recipients who were in the various age groups under 60, there were improved short-term and long-term outcomes following bilateral lung transplantation compared to single. However, for patients 60 years and older, this was reversed. Specifically, at 30 days the mortality was 22 percent for those over 60 following bilateral lung transplant versus 7 percent for those over 60 following single lung transplant. While our study does not specifically show that single lung transplant is better than bilateral in patients 60 and over, it has been our practice to do so, and with this we have favorable outcomes.
Regarding the lack of a statistical difference in hospital mortality between the groups, it certainly is possible that with a larger sample size, we may have seen a significant difference.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Mahidhara, S. Bastani, D. J. Ross, R. Saggar, J. Lynch III, G. T. Schnickel, D. Gjertson, R. Beygui, and A. Ardehali Lung transplantation in older patients? J. Thorac. Cardiovasc. Surg., February 1, 2008; 135(2): 412 - 420. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. F. Rabe, B. Beghe, F. Luppi, and L. M. Fabbri Update in Chronic Obstructive Pulmonary Disease 2006 Am. J. Respir. Crit. Care Med., June 15, 2007; 175(12): 1222 - 1232. [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 |