This Article Abstract Full Text (PDF) All Versions of this Article: 200202-087OCv1 166/12/1550    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mayer-Hamblett, N. Articles by Aitken, M. L. Search for Related Content PubMed PubMed Citation Articles by Mayer-Hamblett, N. Articles by Aitken, M. L.

Developing Cystic Fibrosis Lung Transplant Referral Criteria Using Predictors of 2-Year Mortality

Departments of Pediatrics and Medicine, University of Washington; and Statistical Analysis Unit, Cystic Fibrosis Therapeutics Development Network Coordinating Center, Children's Hospital and Regional Medical Center, Seattle, Washington

Correspondence and requests for reprints should be addressed to Nicole Mayer-Hamblett, Ph.D., Department of Pediatrics, University of Washington, 4800 Sand Point Way N.E., Box 5371, CL-11, Seattle, WA 98105-0371. E-mail: nhambl{at}chmc.org

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
The first objective of our study was to develop a model identifying the best clinical predictors of 2-year mortality among patients with cystic fibrosis (CF), to assist in selection of appropriate candidates for lung transplantation. Using multivariate logistic regression, we found that age, height, FEV₁, respiratory microbiology, number of hospitalizations for pulmonary exacerbations, and number of home intravenous antibiotic courses were all significant predictors of 2-year mortality among 14,572 patients in the Cystic Fibrosis Foundation National Patient Registry who were 6 years of age or older in 1996. The second objective was to compare the diagnostic accuracy of our model when used to guide referral for lung transplant with that of the widely used criterion of an FEV₁ of less than 30% predicted. Surprisingly, this well-fitting model derived from the largest collection of data available on patients with CF provided no better diagnostic accuracy than the simpler FEV₁ criterion. Both had high negative predictive values (98 and 97%, respectively) but only modest positive predictive values (33 and 28%, respectively). Transplant referral decisions based either on a multivariate logistic model or on the criterion of an FEV₁ of less than 30% predicted are likely to result in high rates of premature referral. Better clinical predictors of short-term mortality among patients with CF are needed.

Key Words: cystic fibrosis • mortality • lung transplantation • logistic models

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Over the past 30 years, the median age of survival in patients with cystic fibrosis (CF) in the U.S. has increased from 14 years in 1969 to 32 years in 1999 (1). Nevertheless, approximately 80% of patients with CF still die prematurely due to end-stage lung disease (1). When conventional medical regimens fail, a final treatment option for some patients with CF is lung transplantation. CF is currently the second leading indication for bilateral lung transplantation (2), with 114 to 136 transplants having been performed annually in patients with CF between 1996 and 1999 (1).

Although the choice of appropriate candidates and the timing of referral are critical to the success of lung transplantation for CF, the criteria with which to guide these decisions remain controversial. Because of a shortage of donor organs, wait times between listing for transplantation and receiving lungs are routinely up to 2 years (2), so that delayed referral may result in death before transplant. Indeed, 15 to 40% of listed patients with CF currently die awaiting transplant (3). Conversely, lung transplantation carries significant inherent risks, including a 40% 3-year mortality rate (4), so that premature referral for transplantation could shorten life expectancy.

Current U.S. and European guidelines for CF transplantation recommend an FEV₁ of less than 30% predicted in conjunction with other clinical indicators, such as rapidly progressive respiratory deterioration, hypercarbia, hypoxia, and female sex, as criteria with which to identify patients with CF potentially within the transplant window (4–6). These guidelines are based on several models of short-term mortality among patients with CF that used data from a small number of subjects at a single center (7–11), potentially limiting their generalizability. More recently, Liou and coworkers (12) developed a 5-year survivorship model using clinical information from over 11,000 patients from the Cystic Fibrosis Foundation (CFF) National Patient Registry in the U.S. However, no large, multicenter studies have been conducted identifying predictors of 2-year mortality (a realistic wait time for transplant after listing).

Predictive models for mortality use a patient's clinical characteristics to predict how likely he or she is to die within the specified time period. The patient's probability of dying can then be compared with a cutoff point. If the patient's probability of dying exceeds this cutoff point, he or she may be an appropriate candidate for transplant referral. Thus, the continuous outcome generated by the model (the probability of dying) is converted to a yes/no outcome (the patient should or should not be considered for transplant). Clinicians and patients must know the sensitivity, specificity, and most importantly, the positive and negative predictive values associated with the chosen cutoff point for using the model in an informed manner. How frequently does applying the model produce a false-positive result (leading to referral of a patient for transplant too early) or a false-negative result (leading to referral too late)?

Previous investigators of short-term mortality among patients with CF have not evaluated the accuracy of their models as diagnostic tools relative to different cutoff points. These investigators have evaluated the goodness of fit of their models, which only assesses how closely the model fits the observed data. In other words, do the expected probabilities of dying based on the model match well with the observed probabilities of dying within groups of patients with similar characteristics? Unfortunately, there is not necessarily a correlation between the goodness of fit of a model and its accuracy as a diagnostic tool for classifying patients into those likely to survive and those likely to die (13).

The first objective of our study was to develop a model of 2-year mortality among patients with CF in the U.S., overcoming the limitations of prior studies by using recent data from a large population-based cohort of patients in the CFF National Patient Registry, validating the model in an independent cohort, and investigating a time period relevant to the time a patient might await transplantation. The second objective was to assess the diagnostic accuracy of the model as a tool for identifying patients for lung transplant referral. The trade-offs between referral too early or too late are quantified in terms of sensitivity, specificity, and positive and negative predictive values. The diagnostic accuracy of our model is compared with that of the current widely accepted criterion of an FEV₁ of less than 30% predicted.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Subjects and Data Collection
Subjects were all patients in the CFF National Patient Registry (14) who were 6 years of age or older on December 31, 1996, had not previously undergone lung transplant, and were seen at a CFF-accredited care center in 1996. Use of CFF Registry data was reviewed and approved by the CFF and the Institutional Review Board at the Children's Hospital and Regional Medical Center, Seattle, WA.

For each patient, the following variables were extracted from 1996 Registry data: age on December 31, 1996, sex, CF genotype, race, height, weight, FEV₁, Pseudomonas aeruginosa and Burkholderia cepacia respiratory culture status, number of home intravenous antibiotic courses, and number of hospitalizations for acute exacerbations. Height, weight, and FEV₁ were expressed as the mean of up to four quarterly measurements. Rate of decline in FEV₁ in the 3 years before December 31, 1996, was also considered as a predictor, described for each patient by using an intercept and linear slope term fitted from at least two FEV₁ data points recorded in the 3 years preceding study entry. To minimize bias, potential predictors of survival were excluded if the frequency of missing data was high or if standardized criteria for establishing the characteristic/diagnosis were lacking (i.e., pancreatic insufficiency and diabetes mellitus). Rare characteristics were also excluded. The 2-year vital status outcome for each patient was obtained using 1997 and 1998 Registry data.

Model Development
Univariate logistic regression was used to determine significant independent predictors of 2-year mortality. Multiple logistic regression was then used to assess the effect of all the variables in the same model on 2-year mortality. A full model including all potential predictors was initially fit to the data using a generalized additive model to evaluate the relationship of each continuous predictor with the response (15). The parametric functional forms of the continuous predictors suggested by this model were then considered as alternative predictor variables. In some instances, continuous predictors were converted to categoric counterparts. Stepwise model selection was used to select among all possible predictors and their two-way interactions using goodness-of-fit tests. The final model was then evaluated using standard diagnostic methods to assess the sensitivity of the model to outlying observations (15).

Model Validation and Assessment of Diagnostic Accuracy
A randomly chosen subset of 90% of the study cohort was used for constructing the multiple logistic regression model. The remaining 10% of subjects served as the validation cohort. The model was validated by assessing its goodness of fit in the validation cohort using the Hosmer–Lemeshow test (13). The sensitivity, specificity, and positive and negative predictive values of the final model were calculated by comparing the observed vital status outcomes of patients in the validation cohort with those predicted from the model based on different cutoff values for the probability of dying within 2 years. Similar calculations were performed for various FEV₁ cutoff points. Receiver operator characteristic curves were used to display sensitivity versus 1-specificity for different mortality probability and FEV₁% predicted cutoff points (16).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Cohort Characteristics
There were 14,572 patients in the CFF National Patient Registry in 1996 who met the inclusion criteria. Of these patients, 637 (4%) died within 2 years, 13,021 (90%) were alive at the end of 2 years, and 914 (6%) were lost to follow-up. Of the 637 deaths, 541 (85%) were due to cardiorespiratory disease. The mean age at death was 25.6 years (SD, 10.4 years). Two hundred seventy-eight patients (2% of the cohort) received lung transplants during the 2-year observation period. Of the patients who received transplants, 59 (21%) died during the follow-up period, and 85% of these deaths were considered transplant-related.

A descriptive analysis of the study cohort showed that patients whose 2-year vital status outcome was unknown (n = 914, 6%) did not appear to be inherently different in terms of demographic and clinical characteristics from patients who were alive at the end of the 2-year follow-up. It is estimated that approximately 95% of all deaths due to CF are captured in the CFF Registry (14). Therefore, it was reasonable to assume that patients whose 2-year vital status outcomes were unknown were actually alive at the end of the 2-year observation period. Because removing these patients from the analysis altogether could bias the results, these patients were instead conservatively treated as alive for the purpose of the analysis.

Table 1 demonstrates the pronounced differences between the baseline characteristics of patients who died during the 2-year follow-up period and those of patients who survived. Patients who died were on average older and had lower FEV₁% predicted, lower weight and height percentiles, higher rates of respiratory colonization with B. cepacia and P. aeruginosa, and higher rates of hospitalization for acute pulmonary exacerbations and treatment with home intravenous antibiotics.

Multiple Logistic Regression Model for Predicting 2-Year Mortality
The final multiple logistic regression model is displayed in Table 2 . Age in years was found to be piecewise linearly related to 2-year mortality with a change point at 21 years of age, signifying different rates in the odds of death between patients aged 21 years or less and patients aged more than 21 years. The odds of death associated with a 5-year difference in age was estimated to be 1.32 (95% confidence interval [CI]: 1.28, 1.36) for those aged 21 years or less and 1.12 (95% CI: 1.11, 1.14) for those aged more than 21 years. Patients who had respiratory cultures positive for both P. aeruginosa and B. cepacia were estimated to have the greatest odds of dying within 2 years, compared with those from whom cultures were obtained but who were not colonized with either pathogen (odds ratio [OR]: 4.1; 95% CI: 2.4, 6.9). Patients with B. cepacia alone had a much greater odds of dying within 2 years (OR: 3.23; 95% CI: 1.75, 5.98) than those with P. aeruginosa alone (OR: 1.62; 95% CI: 1.11, 2.35). Patients who were hospitalized for acute exacerbations more than twice annually or had more than two courses of home intravenous antibiotics annually had a significantly greater odds of dying within 2 years compared with those who had no hospitalizations or home intravenous antibiotic courses (OR: 3.53 and 1.49, respectively). Within the context of this model, for each centimeter increase in height, there was a 4% increased odds of dying within 2 years even after adjustment for age (95% CI: 3%, 5%). For each liter increase in FEV₁, the odds of dying within 2 years significantly decreased by 9% (95% CI: 7%, 11%).

Validation of the Model for Goodness of Fit
The ability of the final model to predict 2-year mortality was externally validated and compared with the predictive ability of FEV₁% predicted in a randomly chosen subset of 10% of the study cohort (n = 1,400; 64 [4.6%] died). The Hosmer–Lemeshow goodness-of-fit test indicated that there were no significant differences between the model predictions and the observed data (p = 0.19).

Assessment of Diagnostic Accuracy
Sensitivity and specificity for various cutoff points of model probability and FEV₁ are demonstrated in the receiver operator characteristic curves in Figure 1 . The receiver operator characteristic curves overlap considerably, indicating that the diagnostic accuracy of our model is not significantly better than that of the FEV₁ criterion. This finding is also illustrated in Table 3 , which provides a summary of the diagnostic accuracy of both the model and FEV₁ criterion at various cutoff points. To facilitate comparison between the model and FEV₁ criterion, specificity was fixed for each classification system at four levels (99, 95, 90, and 85%), and the corresponding sensitivity and predictive values were calculated. Both the model and the FEV₁ criterion had high negative predictive values but only modest positive predictive values. Thus, both approaches were better at predicting who would survive 2 years than who would die.

View larger version (18K): [in this window] [in a new window]   Figure 1. Receiver operator characteristic curves for both the final multiple logistic regression model, constructed using different probability-of-dying cutoff values (solid line), and for the FEV1% predicted criterion, constructed using different FEV1% predicted cutoff values (dotted line).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Age, respiratory microbiology, height, FEV₁, and annual number of hospital admissions and courses of home intravenous antibiotics for pulmonary exacerbations were found to be the most important predictors of 2-year mortality in our model. In general, our model agrees well with prior models of short-term mortality in CF (7–12). Liou and coworkers (12) examined predictors of 5-year mortality in a similar cohort of patients, also using CFF National Patient Registry data. The predictors in their final model differ slightly from ours most likely because we analyzed predictors of 2-year rather than 5-year mortality and chose to exclude predictors that are rare or for which the diagnosis is not standardized. For example, they found sex to be a significant predictor of 5-year mortality, whereas we did not find it to be a significant predictor of 2-year mortality. Previous studies have suggested that different models might need to be developed for pediatric and adult patients (17). We found age to be an important predictor of 2-year mortality, with patients aged 21 years or less having a higher risk of death for each increasing year of age than patients aged more than 21 years. However, we found no significant age interactions, and therefore a single model was developed to describe both pediatric and adult patients. Prior studies have also suggested that the rate of decline in FEV₁ may be a stronger predictor of mortality than a threshold or baseline value (8, 17). In our study, as in the study by Liou and coworkers (12), the rate of decline in FEV₁ in the 3 years preceding study entry did not add significantly to a model already containing baseline FEV₁.

It should be noted that our model does not address risk factors for poorer outcomes in the posttransplant period. For example, infection with B. cepacia, particularly genomovar III (18), has clearly been identified as a risk factor for greater mortality after lung transplant (18, 19). Because our goal was to identify a set of risk factors for poor short-term survival in patients who have not yet undergone transplant, we do not attempt to address questions of which risk factors should be seen as relative or absolute contraindications for transplant.

Our use of the CFF National Patient Registry limited our choice of potential predictors of mortality to those contained in the Registry. We were unable to assess clinical variables not contained in the Registry such as oxygen saturations, PA_O2, PA_CO2, heart rate, exercise testing, and hemoglobin that have been shown in prior studies to be important predictors of mortality. For variables contained in the Registry, we were also limited to the frequency of data collection and level of detail provided by the Registry. Perhaps, access to more frequent pulmonary function measurements for each individual would have enabled us to calculate the rate of decline in FEV₁ with greater precision so that it might have become a significant contributor to the model. Lastly, our model may not be generalizable to patients with CF in centers not represented in the Registry.

The relatively low positive predictive value of our model reflects, in part, the low mortality rate of the cohort and also indicates that there may be important unmeasured predictors of death not accounted for in the model. Vizza and coworkers (20) reached conclusions similar to ours in their study attempting to identify risk factors for death among patients with CF already listed for transplant. They were unable to identify a group of clinical or physiologic parameters that reliably differentiated between those who died during the average 2-year wait-list period and those who survived, despite having access to specialized clinical data, such as 6-minute walk distance and pulmonary artery pressure, not available in the CF Registry. The poor ability of clinical characteristics to adequately predict short-term survival among patients with CF contrasts with other disease settings such as idiopathic pulmonary fibrosis, for which 2-year mortality models with high diagnostic accuracy have been developed (21, 22). Unfortunately, at least part of the reason for the greater accuracy of predictive models in idiopathic pulmonary fibrosis is the high mortality rate associated with that condition, with a median survival of just 2.8 years after diagnosis (21).

Recently, Liou and coworkers (23) showed that their 5-year survivorship model was able to identify a high-risk group of patients with CF undergoing lung transplantation who gained a survival benefit from transplantation. However, these investigators did not address the diagnostic accuracy of their model when used as a tool for identifying patients who would gain a survival benefit from transplant, and more importantly, they did not attempt to identify patients to be considered for listing for lung transplant. The cohort of transplant recipients in their analysis included only patients who survived the period from listing to transplant. The best criteria with which to identify patients with CF appropriate for lung transplant referral thus remain unclear.

The high rates of premature referral associated with existing guidelines for transplant referral are responsible for some of the weaknesses of the current waiting system in North America. Patients with an FEV₁ of less than 30% predicted may survive without a transplant, but we have trouble identifying which high-risk patients these may be. Thus, many listed patients do not actually undergo transplant when an organ becomes available. If models predicting short-term mortality are to be truly useful, their diagnostic accuracy will need to be improved.

Perhaps, a two-tier approach would be fruitful. First, a model that uses readily available clinical characteristics could be used to identify high-risk patients. These patients could then undergo more extensive and specialized evaluations such as 6-minute walk testing, echocardiography or cardiac catheterization for pulmonary artery pressures (19), and measurement of hemoglobin, PA_O2, and PA_CO2. The results of these "second tier" tests in these high-risk individuals might then be used to develop a model that better discriminates those likely to survive from those likely to die. Specific probability cutoff points could be established on the basis of the optimal points on the receiver operator characteristic curve for discriminating between survivors and nonsurvivors (21). Perhaps, a web-based tool could be established for ease of use of the model in clinical practice (12).

In the meantime, any guidelines for the optimal timing of lung transplant referral should be used with an understanding of their limitations and accuracy. Use of the FEV₁ criterion or a model such as ours is likely to result in a high false-positive rate, i.e., a large number of patients referred prematurely for transplant. Clinicians and patients must therefore be fully informed of these rates when deciding if and when lung transplant referral is appropriate. The final decision to proceed toward transplant must derive from a comprehensive evaluation of each patient's medical and psychosocial status as well as quality of life.

	Acknowledgments

 
The authors thank their colleagues Michal Kulich, Ph.D., Richard A. Kronmal, Ph.D., and Preston Campbell III, M.D., for their valuable comments, Monica Brooks, B.S., for technical assistance with the CFF National Patient Registry, and the patients with CF who made this study possible.

	FOOTNOTES

 
Supported by the Cystic Fibrosis Foundation and NIH K23 RR15529.

Received in original form February 7, 2002; accepted in final form August 9, 2002

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 

1. Cystic Fibrosis Foundation. Patient Registry 1999 Annual Data Report. Bethesda, MD: Cystic Fibrosis Foundation; 2000.
2. United Network for Organ Sharing. 1999 Annual Report of the U.S. Scientific Registry for Transplant Recipients and the Organ Procurement and Transplantation Network: Transplant Data: 1989–1998. 1999. Rockville, MD: U.S. Department of Health and Human Services, Health Resources and Services Administration, Office of Special Programs, Division of Transplantation; 1999.
3. Flume P. Cystic fibrosis: when to consider lung transplantation? Chest 1998;113:1159–1161.[Free Full Text]
4. Yankaskas JR, Mallory GB, and Consensus Committee. Lung transplantation in cystic fibrosis: consensus conference statement. Chest 1998;113:217–226.[Abstract/Free Full Text]
5. American Society for Transplant Physicians. American Thoracic Society, European Respiratory Society, and International Society for Heart and Lung Transplantation: international guidelines for the selection of lung transplant candidates. Am J Respir Crit Care Med 1998;158:335–339.[Free Full Text]
6. Kotloff RM, Zuckerman JB. Lung transplantation for cystic fibrosis: special considerations. Chest 1996;109:787–798.[Free Full Text]
7. Kerem E, Reisman J, Corey M, Canny GJ, Levison H. Prediction of mortality in patients with cystic fibrosis. N Engl J Med 1992;326:1187–1191.[Abstract]
8. Milla CE, Warwick WJ. Risk of death in cystic fibrosis patients with severely compromised lung function. Chest 1998;113:1230–1234.[Abstract/Free Full Text]
9. Knoke JD, Stern RC, Doershuk CF, Boat TF, Matthews LW. Cystic fibrosis: the prognosis for five-year survival. Pediatr Res 1978;12:676–679.[Medline]
10. Hayllar KM, Williams SG, Wise AE, Pouria S, Lombard M, Hodson ME, Westaby D. A prognostic model for the prediction of survival in cystic fibrosis. Thorax 1997;52:313–317.[Abstract]
11. Aurora P, Wade A, Whitmore P, Whitehead B. A model for predicting life expectancy of children with cystic fibrosis. Eur Respir J 2000;16:1056–1060.[Abstract]
12. Liou TG, Adler FR, Fitzsimmons SC, Cahill BC, Hibbs JR, Marshall BC. Predictive 5-year survivorship model of cystic fibrosis. Am J Epidemiol 2001;153:345–352.[Abstract/Free Full Text]
13. Hosmer DW, Lemeshow S. Applied logistic regression. New York: Wiley; 1989.
14. Fitzsimmons SC. The changing epidemiology of cystic fibrosis. J Pediatr 1993;122:1–9.[Medline]
15. McCullagh P, Nelder JA. Generalized linear models. London: Chapman and Hall; 1983.
16. Metz CF. Basic principles of ROC analysis. Semin Nucl Med 1978;8:283–298.[Medline]
17. Robinson W, Waltz DA. FEV(1) as a guide to lung transplant referral in young patients with cystic fibrosis. Pediatr Pulmonol 2000;30:198–202.[CrossRef][Medline]
18. Aris RM, Routh JC, KiPuma JJ, Heath DG, Gilligan PH. Lung transplantation for cystic fibrosis patients with Burkholderia cepacia complex: survival linked to genomovar type. Am J Respir Crit Care Med 2001;164:2102–2106.[Abstract/Free Full Text]
19. Chaparro C, Maurer J, Gutierrez C, Krajden M, Chan C, Winton T, Keshavjee S, Scavuzzo M, Tullis E, Hutcheon M, et al. Infection with Burkholderia cepacia in cystic fibrosis: outcome following lung transplantation. Am J Respir Crit Care Med 2001;163:43–48.[Abstract/Free Full Text]
20. Vizza CD, Yusen RD, Lynch JP, Fedele F, Patterson GA, Trulock EP. Outcome of patients with cystic fibrosis awaiting lung transplantation. Am J Respir Crit Care Med 2000;162:819–825.[Abstract/Free Full Text]
21. Mogulkoc N, Brutsche MH, Bishop PW, Greaves SM, Horrocks AW, Egan JJ. Pulmonary function in idiopathic pulmonary fibrosis and referral for lung transplantation. Am J Respir Crit Care Med 2001;164:103–108.[Abstract/Free Full Text]
22. King TE Jr, Tooze JA, Schwarz MI, Brown KR, Cheriack RM. Predicting survival in idiopathic pulmonary fibrosis: scoring system and survival model. Am J Respir Crit Care Med 2001;164:1171–1181.[Abstract/Free Full Text]
23. Liou TG, Adler FR, Cahill BC, Fitzsimmons SC, Huang D, Hibbs JR, Marshall BC. Survival effect of lung transplantation among patients with cystic fibrosis. JAMA 2001;286:2683–2689.[Abstract/Free Full Text]

This article has been cited by other articles:

				J W Wilson, R M du Bois, and T E King Jr Challenges in pulmonary fibrosis: 8 {middle dot} The need for an international registry for idiopathic pulmonary fibrosis Thorax, March 1, 2008; 63(3): 285 - 287. [Abstract] [Full Text] [PDF]

				M. Stern, B. Wiedemann, P. Wenzlaff, and on behalf of the German Cystic Fibrosis Quality As From registry to quality management: the German Cystic Fibrosis Quality Assessment project 1995 2006 Eur. Respir. J., January 1, 2008; 31(1): 29 - 35. [Abstract] [Full Text] [PDF]

				M. P. Boyle Adult Cystic Fibrosis JAMA, October 17, 2007; 298(15): 1787 - 1793. [Abstract] [Full Text] [PDF]

				N. Mayer-Hamblett, B. W. Ramsey, and R. A. Kronmal Advancing Outcome Measures for the New Era of Drug Development in Cystic Fibrosis Proceedings of the ATS, August 1, 2007; 4(4): 370 - 377. [Abstract] [Full Text] [PDF]

				C. H. Goss and A. L. Quittner Patient-reported Outcomes in Cystic Fibrosis Proceedings of the ATS, August 1, 2007; 4(4): 378 - 386. [Abstract] [Full Text] [PDF]

				N. Mayer-Hamblett, M. L. Aitken, F. J. Accurso, R. A. Kronmal, M. W. Konstan, J. L. Burns, S. D. Sagel, and B. W. Ramsey Association between Pulmonary Function and Sputum Biomarkers in Cystic Fibrosis Am. J. Respir. Crit. Care Med., April 15, 2007; 175(8): 822 - 828. [Abstract] [Full Text] [PDF]

				D. Hadjiliadis Special Considerations for Patients With Cystic Fibrosis Undergoing Lung Transplantation Chest, April 1, 2007; 131(4): 1224 - 1231. [Abstract] [Full Text] [PDF]

				C. H Goss and J. L Burns Exacerbations in cystic fibrosis {middle dot} 1: Epidemiology and pathogenesis Thorax, April 1, 2007; 62(4): 360 - 367. [Abstract] [Full Text] [PDF]

				K. Buranawuti, M. P Boyle, S. Cheng, L. L Steiner, K. McDougal, M D. Fallin, C. Merlo, P. L Zeitlin, B. J Rosenstein, P. J Mogayzel Jr, et al. Variants in mannose-binding lectin and tumour necrosis factor {alpha} affect survival in cystic fibrosis J. Med. Genet., March 1, 2007; 44(3): 209 - 214. [Abstract] [Full Text] [PDF]

				E. F. McKone, C. H. Goss, and M. L. Aitken CFTR Genotype as a Predictor of Prognosis in Cystic Fibrosis. Chest, November 1, 2006; 130(5): 1441 - 1447. [Abstract] [Full Text] [PDF]

				J K Block, K L Vandemheen, E Tullis, D Fergusson, S Doucette, D Haase, Y Berthiaume, N Brown, P Wilcox, P Bye, et al. Predictors of pulmonary exacerbations in patients with cystic fibrosis infected with multi-resistant bacteria Thorax, November 1, 2006; 61(11): 969 - 974. [Abstract] [Full Text] [PDF]

				R. A. Belkin, N. R. Henig, L. G. Singer, C. Chaparro, R. C. Rubenstein, S. X. Xie, J. Y. Yee, R. M. Kotloff, D. A. Lipson, and G. R. Bunin Risk Factors for Death of Patients with Cystic Fibrosis Awaiting Lung Transplantation Am. J. Respir. Crit. Care Med., March 15, 2006; 173(6): 659 - 666. [Abstract] [Full Text] [PDF]

				C. H. Goss, G. D. Rubenfeld, B. W. Ramsey, and M. L. Aitken Clinical Trial Participants Compared with Nonparticipants in Cystic Fibrosis Am. J. Respir. Crit. Care Med., January 1, 2006; 173(1): 98 - 104. [Abstract] [Full Text] [PDF]

				P. A. Flume, C. Strange, X. Ye, M. Ebeling, T. Hulsey, and L. L. Clark Pneumothorax in Cystic Fibrosis Chest, August 1, 2005; 128(2): 720 - 728. [Abstract] [Full Text] [PDF]

				T. M. Egan and R. M. Kotloff Pro/Con Debate: Lung Allocation Should Be Based on Medical Urgency and Transplant Survival and Not on Waiting Time* Chest, July 1, 2005; 128(1): 407 - 415. [Full Text] [PDF]

				S. D. Nathan Lung Transplantation: Disease-Specific Considerations for Referral Chest, March 1, 2005; 127(3): 1006 - 1016. [Abstract] [Full Text] [PDF]

				R. Kraemer, A. Blum, A. Schibler, R. A. Ammann, and S. Gallati Ventilation Inhomogeneities in Relation to Standard Lung Function in Patients with Cystic Fibrosis Am. J. Respir. Crit. Care Med., February 15, 2005; 171(4): 371 - 378. [Abstract] [Full Text] [PDF]

				D. B. Rosenbluth, K. Wilson, T. Ferkol, and D. P. Schuster Lung Function Decline in Cystic Fibrosis Patients and Timing for Lung Transplantation Referral Chest, August 1, 2004; 126(2): 412 - 419. [Abstract] [Full Text] [PDF]

				A.R. Glanville and M. Estenne Indications, patient selection and timing of referral for lung transplantation Eur. Respir. J., November 1, 2003; 22(5): 845 - 852. [Abstract] [Full Text] [PDF]

				R. L. Gibson, J. L. Burns, and B. W. Ramsey Pathophysiology and Management of Pulmonary Infections in Cystic Fibrosis Am. J. Respir. Crit. Care Med., October 15, 2003; 168(8): 918 - 951. [Abstract] [Full Text] [PDF]

				M. J. Tobin Pediatrics, Surfactant, and Cystic Fibrosis in AJRCCM 2002 Am. J. Respir. Crit. Care Med., February 1, 2003; 167(3): 333 - 344. [Full Text] [PDF]

				M. J. Tobin Chronic Obstructive Pulmonary Disease, Pollution, Pulmonary Vascular Disease, Transplantation, Pleural Disease, and Lung Cancer in AJRCCM 2002 Am. J. Respir. Crit. Care Med., February 1, 2003; 167(3): 356 - 370. [Full Text] [PDF]

				P. G. Noone and T. M. Egan Cystic Fibrosis: When to Refer for Lung Transplantation-Is the Answer Clear? Am. J. Respir. Crit. Care Med., December 15, 2002; 166(12): 1531 - 1532. [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 200202-087OCv1 166/12/1550    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mayer-Hamblett, N. Articles by Aitken, M. L. Search for Related Content PubMed PubMed Citation Articles by Mayer-Hamblett, N. Articles by Aitken, M. L.