Ann Thorac Surg 2009;88:1757-1764. doi:10.1016/j.athoracsur.2009.07.005
© 2009 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Lung Allocation Score Predicts Survival in Lung Transplantation Patients With Pulmonary Fibrosis
Eric S. Weiss, MD, MPHa,
Jeremiah G. Allen, MDa,
Christian A. Merlo, MD, MPHb,c,
John V. Conte, MDa,
Ashish S. Shah, MDa,*
a Division of Cardiac Surgery, The Johns Hopkins Medical Institution, Baltimore, Maryland
b Division of Pulmonary and Critical Care Medicine, The Johns Hopkins Medical Institution, Baltimore, Maryland
c Bloomberg School of Public Health, The Johns Hopkins Medical Institution, Baltimore, Maryland
Accepted for publication July 2, 2009.
* Address correspondence to Dr Shah, Department of Surgery, Division of Cardiac Surgery, The Johns Hopkins Hospital, Blalock 618, 600 N Wolfe St, Baltimore, MD 21287 (Email: ashah29{at}jhmi.edu).
Presented at the Poster Session of the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26–28, 2009.
 |
Abstract
|
|---|
Background: Since 2005, the Organ Procurement and Transplantation Network has used the lung allocation score (LAS) to assign organ allocation priority in lung transplantation. This study was designed to determine whether LAS predicts short-term survival for patients with pulmonary fibrosis.
Methods: Organ Procurement and Transplantation Network data was retrospectively reviewed to identify 1,256 first-time adult lung transplantation recipients with pulmonary fibrosis since initiation of the LAS (May 2005 to December 2007). Patients were stratified by quartiles of LAS. Multivariable Cox proportional hazards regression predicted the risk of 30-day, 90-day, and 1-year mortality.
Results: Lung allocation scores ranged from 31.1 to 94.1. Lung allocation score quartiles (Q) were as follows: Q1, 29.8 to 37.8, n = 315; Q2, 37.9 to 42.5, n = 313; Q3, 42.6 to 51.9, n = 314; and Q4, 52.0 to 94.1, n = 314. Lung allocation score correlated strongly with cumulative survival at 90 days and 1 year after lung transplantation. Patients in the highest LAS quartile (LAS Q4, 52.0 to 94.1) had a 10% lower cumulative survival at 1 year after transplantation when compared with patients in the lowest LAS quartile (p = 0.04). On Cox proportional hazards regression, patients in the highest LAS quartile (those above 52.0) had a significant increase in the risk of mortality at both 90 days and 1 year after transplantation (relative to reference Q1: hazard ratio, 2.09; 95% confidence interval, 1.16 to 3.80; p = 0.01 for 90 days; and hazard ratio, 1.59; 95% confidence interval, 1.04 to 2.44; p = 0.03 for 1 year).
Conclusions: An initial examination of survival for pulmonary fibrosis lung transplantation recipients in the post-LAS era was performed. Lung allocation score predicts short-term mortality in this cohort.
|
Introduction
|
|---|
Although lung transplantation (LTx) has been shown to improve survival and quality of life [1, 2], it is an imperfect therapy constrained in part by limitations in donor organ supply. In May 2005, the United Network for Organ Sharing implemented the lung allocation scoring (LAS) system to prioritize patients on the waiting list. The LAS is a standardized numeric score ranging from 0 to 100 points that seeks to identify patients likely to benefit most from LTx. The score is derived from the difference between a recipient's expected 1-year posttransplant survival and the likelihood of dying on the waiting list (Table 1) [3–5].
Implementation of the LAS has dramatically changed the demographics of LTx. High LAS candidates are allocated organs rapidly and preferentially. Consequently, patients with idiopathic pulmonary fibrosis (IPF), an interstitial lung disease leading to acute respiratory failure and death, are being allocated donor lungs with increased frequency [6]. Because of the nature of their disease process, IPF patients frequently present with heightened clinical acuity and, consequently, have lower overall posttransplant survival when compared with patients with other disease types [6–8].
Successful organ allocation demands an often-delicate balance between identifying those patients most in need and providing organs to those likely to realize maximum benefit. It was with this purpose in mind that the LAS was implemented. In addition to identifying patients at high risk for death on the waiting list [9], the LAS may provide a means for risk stratification in IPF. However, it remains unclear whether the LAS predicts posttransplant survival for these patients. Identification of subgroups of patients with exceedingly high risk might enable the transplant community to predict long-term prognosis and identify patients at prohibitive risk. To this end, we examined United Network for Organ Sharing records to identify a cohort of IPF patients receiving LTx after the implementation of the LAS system, to examine short-term survival as a function of recipient LAS. We hypothesize that for patients with IPF, LAS will inversely correlate with post-LTx short-term survival.
|
Material and Methods
|
|---|
Data Source
The dataset used (United Network for Organ Sharing Standard Transplant Analysis and Research files) represents a prospectively collected open cohort of all U.S. patients receiving LTx from 1987 until May 2008. No patient or center identifiers were included in the analysis, and the study was reviewed and granted institutional review board exemption at our institution.
Study Design
We conducted a retrospective cohort study, examining adult (>17 years) patients with pulmonary fibrosis (PF; coded as idiopathic or other) who received LTx during an interval associated with implementation of the LAS system in the United States (May 2005 through December 2007). We excluded patients receiving heart-lung transplantation. The primary end point was the incidence of mortality at 1 year after LTx. Also examined was cumulative mortality at 30 and 90 days after LTx. Risk of mortality was examined for LAS as a continuous variable. Additionally, the cohort was stratified by quartiles of LAS.
Variables
Pertinent variables examined included demographic factors (age, sex, race, and education level); markers of pulmonary status (oxygen requirement, 6-minute walking distance, forced expiratory volume at 1 second, forced vital capacity, forced expiratory volume at 1 second to forced vital capacity ratio, mechanical ventilation before LTx, and intensive care unit care before transplantation); comorbidities (diabetes mellitus, body mass index, preoperative creatinine levels, and hypertension); and transplant variables (ischemic time, HLA mismatch, panel-reactive antibody level, year of transplant, and waiting list times). We examined donor variables including age, race, sex, and body mass index. Finally, mean annual center LTx volume (as a continuous variable) was included to adjust for hospital experience.
Statistical Analysis
We compared baseline characteristics among the four LAS quartiles by one-way analysis of variance (continuous variables) and the 
2 test (categorical variables). For significant associations, post hoc pairwise comparisons between strata were performed using the Tukey-Kramer method (continuous variables) and by univariate logistic regression (categorical variables).
Cumulative survival was estimated using the Kaplan-Meier method, focused on time intervals with adequate follow-up. The log-rank test was used to compare survival estimates by group. Risk of death was assessed using a multivariable Cox proportional hazards regression model predicting 1-year mortality with censoring occurring for those individuals lost to follow-up and those alive at the end of study time (administratively censored). Independent covariates not included in the LAS, with potential for confounding based on biologic plausibility, previous literature, or those with probability values less than 0.2 (exploratory level only), were deemed variables of interest. These covariates were incorporated into the Cox model in a forward and backward stepwise fashion using the likelihood ratio test and Akaike's information criterion in a nested model approach to determine which covariates increased the explanatory power of the model. The final risk-adjusted model (model A) incorporated the following covariates: age, recipient race, donor race, ischemic time, and mean annual institutional volume (as a continuous variable). The model was initially applied with LAS as a continuous variable and then applied to LAS quartiles to examine the risk of mortality at 30 days, 90 days, and 1 year after LTx. As all models were constructed by means of casewise deletion, covariates with greater than 15% missing data in the registry were not included.
Sensitivity Analysis
To maximize follow-up time, a sensitivity analysis was conducted (model B) using only those patients undergoing transplantation during the first year of the LAS system (May 1, 2005, to April 30, 2006; median follow-up, 23.5 months; interquartile range, 14.1 to 25.6 months). To determine whether there was a range of LAS within the highest LAS quartile that conferred particularly high risk, Kaplan-Meier estimates of 1-year mortality (with 95% confidence intervals [CI]) were plotted for deciles of LAS. Finally, to determine which LAS components maximally affected posttransplant survival, a Cox proportional hazards regression model predicting the risk of 1-year mortality was conducted incorporating variables used in construction of the waiting list urgency and posttransplant survival components of the LAS.
Data Presentation
For all analyses, a probability value of less than 0.05 (two-tailed) was considered significant. Means are presented with standard deviations, medians with interquartile range, and hazard ratios (HR) with 95% CI. Statistical analyses were performed with the aid of STATA software (version 9.2 Special Edition; StataCorp LP, College Station, TX).
|
Results
|
|---|
Cohort Statistics
A total of 1,275 patients with PF were followed within the United Network for Organ Sharing registry during the study period. After exclusion of children (n = 9), those with previous transplants (n = 8), and those with no follow-up information (n = 2), 1,256 patients made up the final study population. The mean age was 57.7 ± 9.3 years, with 31.2% women (n = 393). Patients spent a median time of 51 days on the waiting list (interquartile range, 15 to 166 days). The median follow-up time was 11.5 months (interquartile range, 2.0 to 18.8).
Stratification
Lung allocation score ranged from 31.1 to 94.1. All patients in the study population had a recorded LAS, and the median score was 42.9 (intraquartile range, 37.8 to 52.0). Stratification by quartile was as follows: LAS quartile 1, 29.8 to 37.8, n = 315; quartile 2, 37.9 to 42.5, n = 313; quartile 3, 42.6 to 51.9, n = 314; and quartile 4, 52.0 to 94.1, n = 314.
Baseline Characteristics
Patients in the lowest LAS quartile were younger and had a lower percentage of ethnic minorities (Table 2). Patients in LAS quartile 4 required an average of 7.8 L/min of oxygen at baseline, compared with only 2.7 L/min in quartile 1 (p < 0.001). The percentage of patients walking less than 150 feet in 6 minutes was the highest in LAS quartile 4 (27.4%). Additionally, forced expiratory volume at 1 second and forced vital capacity varied in accordance with recipient LAS. Patients with high LAS were also more likely to be in an intensive care unit and require mechanical ventilation before undergoing transplantation and were more likely to have a history of diabetes mellitus. Importantly, time spent on the waiting list correlated inversely with LAS as those patients in the highest LAS quartile spent the shortest time on the waiting list. Finally, those patients in LAS quartiles 3 and 4 were more likely to receive bilateral lung transplantation than the reference group (LAS quartile 1; Table 2).
Survival
On unadjusted analysis, LAS correlated strongly with cumulative survival at 90 days and 1 year only (Table 3). Thirty-day survival was not significantly affected by LAS. Those patients in LAS quartile 4 (LAS, 52.0 to 94.1) had a 10% lower cumulative survival 1 year after transplantation (73.9%) when compared with patients in the lowest LAS quartile (p = 0.006). Each incremental LAS quartile showed a decreased cumulative survival at 1 year (Fig 1). These differences in survival persisted after censoring for death within the first 30 days (Fig 2). When deciles of LAS were examined, a clear trend toward increasing 1-year mortality with increasing LAS was observed (Fig 3). The highest cumulative mortality occurred for patients with LAS of 70 to 79 (39.5%).
View larger version (22K):
[in this window]
[in a new window]
|
Fig 1. Kaplan-Meier estimates of 1-year survival for pulmonary fibrosis patients stratified by lung allocation score (LAS) quartiles (Q) (Organ Procurement and Transplantation Network data, May 2008).
|
|
View larger version (23K):
[in this window]
[in a new window]
|
Fig 2. Kaplan-Meier estimates of 1-year survival censored for 30-day mortality for pulmonary fibrosis patients by lung allocation score (LAS) quartiles (Q) (from Organ Procurement and Transplantation Network data, May 2008).
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig 3. Kaplan-Meier estimates of 1-year mortality by lung allocation score deciles. Estimate provided with 95% confidence interval. *p < 0.05 for comparison of mortality estimate to reference (quartile 1; Q1) (Organ Procurement and Transplantation Network data, May 2008).
|
|
Multivariable Analysis
Lung allocation score (as a continuous variable) was associated with a significant risk of death within the first year of transplant (HR, 1.02; 95% CI, 1.01 to 1.02; p < 0.001; Table 4). Thus, each increase of one LAS point corresponded to a 2% increase in the risk of mortality at 1 year. When stratified by quartiles, only those patients in the highest LAS quartile (those above 52.0) had a significant increase in the risk of mortality at both 90 days and 1 year after transplantation (relative to reference quartile 1: HR, 2.09; 95% CI, 1.16 to 3.80; p = 0.01 for 90 days and HR, 1.59; 95% CI, 1.04 to 2.44; p = 0.03 for 1 year; Table 5). In this risk-adjusted model, patients in LAS quartiles 2 and 3 did not have an increased risk of mortality at any time relative to quartile 1. On sensitivity analysis maximizing follow-up, patients in quartile 4 continued to manifest an increased risk of 1-year cumulative morality (HR, 2.49; 95% CI, 1.24 to 4.99; p = 0.01; Table 5).
View this table:
[in this window]
[in a new window]
|
Table 5 Cox Proportional Hazard Regression Analysis, Providing Unadjusted and Multivariable Adjusted Risk of Death for Lung Allocation Score Quartiles
|
|
Lung Allocation Score Components
To address which LAS components are predictive of posttransplant survival, two separate breakout multivariable analyses were performed incorporating LAS components used in the waiting list urgency and posttransplant survival models (Table 6). Variables included in the waiting list urgency model that significantly predicted mortality in our series included age (HR, 1.04; 95% CI, 1.01 to 1.09; p = 0.007), body mass index (HR, 0.92; 95% CI, 0.87 to 0.99; p = 0.02), oxygen requirement at rest (HR, 1.04; 95% CI, 1.00 to 1.14; p = 0.04), and mechanical ventilation (HR, 3.26; 95% CI, 1.16 to 9.23; p < 0.001). Significant predictors in the posttransplant survival model included age (HR, 1.04; 95% CI, 1.02–1.07; p < 0.001), creatinine (HR, 1.45; 95% CI, 1.08 to 1.93; p = 0.01), and preoperative mechanical ventilation (HR, 3.78; 95% CI, 2.32 to 6.17; p < 0.001; Table 6).
View this table:
[in this window]
[in a new window]
|
Table 6 Univariate and Multivariable Risk of 1-Year Mortality Associated With Factors Present in the Waiting List Urgency and Posttransplant Survival Models of the Lung Allocation Score
|
|
|
Comment
|
|---|
We have provided early data to show that the LAS predicts 90-day and 1-year mortality for patients with PF. The analysis focused on 1,256 PF patients in the post-LAS era, and demonstrated an incremental effect of LAS on mortality. Risk of mortality was especially high for patients with LAS greater than 52. In a risk-adjusted model, these patients had a doubling of the risk of mortality at 90 days after LTx and a greater than 50% increase in the risk of mortality at 1 year. Increasing age and black donor race were also associated with decreased 1-year survival.
As would be expected, patients with high LAS possess different clinical characteristics from those with low LAS. Patients with high LAS were more likely to be mechanically ventilated before undergoing transplantation, were more likely to have diabetes mellitus, had higher oxygen requirements at baseline, and had a lower percentage of predicted forced expiratory volume at 1 second and forced vital capacity. Importantly, examination of LAS factors predicting mortality revealed that IPF patients being ventilated before transplantation had an exceedingly high risk of 1-year mortality.
Importantly, waiting list times inversely correlated with LAS as IPF patients with high LAS were allocated organs rapidly. In our study, patients with high LAS (quartile 4) spent an average of 69.7 fewer days on the waiting list than those patients in LAS quartile 1. A previous report by Lingaraju and colleagues [10] indicated that the LAS system may have little impact on waiting list times. However, our national sample refutes that assertion. In this regard, use of the LAS does reduce waiting list time for those patients likely to benefit from LTx. A second interesting finding is that LAS varied widely within the sample (31.1 to 94.1). This range represents 63% of the value of the maximum score, indicating that transplantation for IPF occurs across an extremely wide range of clinical acuity.
The Lung Allocation Score
The LAS represents an important paradigm shift in LTx. The previous allocation system used time on the waiting list and perceived urgency as the primary determinants for allocation. In contrast, the LAS quantitatively assigns priority based on the difference of the likelihood of posttransplant survival and waiting list urgency [3–5]. With use of this system, patients of high clinical acuity are rapidly and preferentially allocated organs. This led to a relative increase in patients with IPF receiving donor organs. Our analysis suggests that the LAS can also serve as a quantitative tool to gauge posttransplant survival.
Resource Utilization and Restriction for Patients with High Lung Allocation Score
Given organ scarcity, we must address whether there exists a level of LAS prohibitive for allocation in IPF. A further question is whether posttransplant survival should be given increased weight in the LAS calculation. Certainly, a 1-year cumulative mortality rate of close to 40% for those patients with LAS of 70 to 79 is a sobering statistic. However, it must be remembered that the LAS subtracts posttransplant survival from waiting list urgency and is designed to maximize posttransplant benefit, not posttransplant survival. One-year survival in these high-acuity patients will be higher than the natural history of the disease. We must realize that allocation of lungs to IPF patients offers them the best chance of survival. The ethical, societal, and economic impact of transplanting patients with IPF and high LAS is a subject that requires further debate.
Limitations
Despite our positive findings, our study is not without limitations. The study focused only on those patients with PF. The results of this study cannot be applied to patients with other LTx indications. The retrospective nature of the analysis does not provide control over all potentially confounding variables. The United Network for Organ Sharing dataset provides limited follow-up, and in some cases missing data, and assessment of important parameters beyond survival, such as quality of life and the development of bronchiolitis obliterans, are limited.
Conclusions
In this analysis, we have provided a first look at outcomes for PF patients in the post-LAS era. Lung allocation score predicts mortality 1 year after LTx. Patients with LAS greater than 52 have a 10% decrease in -year posttransplant survival. Further investigations are required to address the societal impact of performing LTx in patients with extremely high LAS.
|
Acknowledgments
|
|---|
Doctor Weiss is the Irene Piccinini Investigator in Cardiac Surgery and Dr Allen is the Hugh R. Sharp Cardiac Surgery Research Fellow. This work was supported in part by Health Resources and Services Administration contract 234-2005-370011C and the National Institutes of Health (NIH 2T32DK007713-12 [ESW]). The content is the responsibility of the authors alone and does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. government.
|
References
|
|---|
- Geertsma A, Ten Vergert EM, Bonsel GJ, et al. Does lung transplantation prolong life?. A comparison of survival with and without transplantation. J Heart Lung Transplant 1998;17:511-516.[Medline]
- TenVergert EM, Essink-Bot ML, Geertsma A, et al. The effect of lung transplantation on health-related quality of life: a longitudinal study Chest 1998;113:358-364.[Abstract/Free Full Text]
- Egan TM, Kotloff RM. Pro/con debate: lung allocation should be based on medical urgency and transplant survival and not on waiting time Chest 2005;128:407-415.[Free Full Text]
- Egan TM, Murray S, Bustami RT, et al. Development of the new lung allocation system in the United States Am J Transplant 2006;6:1212-1227.[Medline]
- United Network for Organ Sharing A guide to calculating the lung allocation scorehttp://www.unos.org/SharedContentDocuments/lung_allocation_score_updated_01072009.pdf 2006Accessed June 5, 2009.
- Christie JD, Edwards LB, Aurora P, et al. Registry of the International Society for Heart and Lung Transplantation: Twenty-Fifth Official Adult Lung and Heart/Lung Transplantation Report—2008 J Heart Lung Transplant 2008;27:957-969.[Medline]
- Nwakanma LU, Simpkins CE, Williams JA, et al. Impact of bilateral versus single lung transplantation on survival in recipients 60 years of age and older: analysis of United Network for Organ Sharing database J Thorac Cardiovasc Surg 2007;133:541-547.[Abstract/Free Full Text]
- Rinaldi M, Sansone F, Boffini M, et al. Single versus double lung transplantation in pulmonary fibrosis: a debated topic Transplant Proc 2008;40:2010-2012.[Medline]
- Merlo CA, Weiss ES, Orens JB, et al. Impact of U.S. Lung Allocation Score on survival after lung transplantation J Heart Lung Transplant 2009;28:769-775.[Medline]
- Lingaraju R, Blumenthal NP, Kotloff RM, et al. Effects of lung allocation score on waiting list rankings and transplant procedures J Heart Lung Transplant 2006;25:1167-1170.[Medline]
This article has been cited by other articles:
|
|
|
|
 
N. C. De Oliveira, S. Osaki, J. Maloney, R. D. Cornwell, and K. C. Meyer
Lung transplant for interstitial lung disease: outcomes for single versus bilateral lung transplantation
Interact CardioVasc Thorac Surg,
December 16, 2011;
(2011)
ivr085v1.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. C. De Oliveira, S. Osaki, J. Maloney, R. D. Cornwell, and K. C. Meyer
Lung transplant for interstitial lung disease: outcomes before and after implementation of the united network for organ sharing lung allocation scoring system
Eur J Cardiothorac Surg,
December 14, 2011;
(2011)
ezr079v1.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. J. George, G. J. Arnaoutakis, and A. S. Shah
Lung Transplant in Idiopathic Pulmonary Fibrosis
Arch Surg,
October 1, 2011;
146(10):
1204 - 1209.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
