| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary and Critical Care Medicine (Drs. Christie and Kotloff), Department of Medicine, University of Pennsylvania School of Medicine; Department of Biostatistics and Epidemiology (Drs. Landis and Kimmel), University of Pennsylvania School of Medicine; Division of Pulmonary and Critical Care Medicine (Dr. Arcasoy), Department of Medicine, Columbia University College of Physicians and Surgeons; and Division of Thoracic Surgery, Department of Surgery (Drs. Pochettino and Rosengard), University of Pennsylvania School of Medicine, Philadelphia, PA.
Correspondence to: Jason D. Christie, MD, MS, Assistant Professor of Medicine and Epidemiology, Division of Pulmonary, Allergy and Critical Care Medicine, University of Pennsylvania School of Medicine, Center for Clinical Epidemiology and Biostatistics, 423 Guardian Dr, 719 Blockley Hall, Philadelphia, PA 19104; e-mail: jchristi{at}cceb.med.upenn.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Design: We conducted a cohort study of 255 consecutive lung transplant procedures performed between October 1991 and July 2000. We defined PGF as follows: (1) diffuse alveolar opacities exclusively involving allograft(s) and developing within 72 h of transplant, (2) a ratio of PaO2 to fraction of inspired oxygen < 200 beyond 48 h postoperatively, and (3) no other secondary cause of graft dysfunction identified. Risk factors were assessed individually and adjusted for confounding using multivariable logistic regression models.
Setting: Tertiary-care academic medical center.
Results: The overall incidence was 11.8% (95% confidence interval [CI], 7.9 to 15.9). Following multivariable analysis, the risk factors independently associated with development of PGF were as follows: a recipient diagnosis of primary pulmonary hypertension (PPH; adjusted odds ratio [OR], 4.52; 95% CI, 1.29 to 15.9; p = 0.018), donor female gender (adjusted OR, 4.11; 95% CI, 1.17 to 14.4; p = 0.027), donor African-American race (adjusted OR, 5.56; 95% CI, 1.57 to 19.8; p = 0.008), and donor age < 21 years (adjusted OR, 4.06; 95% CI, 1.34 to 12.3; p = 0.013) and > 45 years (adjusted OR, 6.79; 95% CI, 1.61 to 28.5; p = 0.009).
Conclusions: Recipient diagnosis of PPH, donor African-American race, donor female gender, and donor age are independently and strongly associated with development of PGF.
Key Words: acute lung injury • gender • lung transplantation • organ donation • pulmonary hypertension • race • reperfusion injury
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Definition of PGF
The definition of PGF is based on criteria previously published by our group3
and represents an adaptation of the American European Consensus Conference definition of ARDS.12
Although there is a spectrum of reperfusion injury following lung transplantation,11
13
we purposefully chose criteria that select the patients with the most severe form of clinical graft dysfunction following transplantation. To be defined as having PGF, study subjects had to meet all of the following criteria: (1) the presence within 72 h of transplantation of a diffuse alveolar infiltrate involving the lung allograft(s) and, in the case of a single-lung transplant, sparing the native lung; (2) ratio of PaO2 to fraction of inspired oxygen (FIO2) of < 200 persisting beyond the initial 48 h postoperatively; (3) no other secondary cause of graft dysfunction identified ([a] cardiogenic pulmonary edema, defined as a pulmonary artery occlusion pressure of > 18 cm or resolution of infiltrates with effective diuresis; [b] pathologic evidence of rejection; [c] pneumonia as evidenced by the presence of fever, leukocytosis, and purulent secretions with positive culture findings on bronchoscopy during the first 3 postoperative days; and [d] pulmonary venous outflow obstruction by clot or kinking as demonstrated by transesophageal echocardiography or direct inspection on surgical re-exploration or postmortem examination); (4) in the event of death prior to day 3, the patient must fulfill the above criteria at the time of death and must demonstrate diffuse alveolar damage as the predominant process on histologic examination of the lung (available on all patients with death within 72 h). All patients in the cohort study with diffuse infiltrates underwent bronchoscopy and transesophageal echocardiography to exclude other causes of graft failure.
Standard Transplant Protocol
Donor selection, graft procurement, immunologic evaluation, surgical technique, postoperative management, and immunosuppression all proceeded according to our standard transplant protocol, which has been previously published.3
Donors were considered suitable if they fulfilled the following criteria: (1) < 60 years old, (2) clear chest radiograph, (3) a PaO2/FIO2 ratio of > 300 while receiving 100% FIO2 and 5 cm H2O positive end-expiratory pressure, and (4) airways free of purulent secretions or gastric contents on direct examination with bronchoscopy. Donor lung preservation utilized 500 µg of prostaglandin E1 and modified Eurocollins solution. Single lung transplantation was performed via a standard posterior thoracotomy approach. The lateral thoracotomy approach was used in few patients during the later years of the study. An anterior thoracosternotomy approach was used for bilateral transplantation. Beginning with the 16th patient in our program, prostaglandin E1 infusion was used in all patients during the first 72 h postoperatively. The drug was not administered to several patients in whom significant hypotension developed. In all cases, methylprednisolone was administered as a 500-mg bolus prior to reperfusion of the allograft and then at a daily dose of 0.5 mg/kg tapered to 0.15 mg/kg by the third postoperative month. Cyclosporine was initiated immediately postoperatively to achieve a whole blood level of 250 to 350 ng/mL. Azathioprine therapy was initiated at a dose of 2 mg/kg/d. All patients were screened prior to surgery for the presence of preformed antibodies to human leukocyte antigen with a panel reactive antibody test, and all were found to have values 
10%.
Candidate Risk Factors for PGF
Risk factors were selected based on hypothesized associations with PGF. Our purpose was to limit the candidate risk factors to those supported by clinical and/or biological plausibility. In general, clinical risk factors were categorized as donor variables, recipient variables, and surgical variables.
Donor risk factors included donor race, gender, age, smoking history, PaO2 following challenge with 100% FIO2 and 5 cm H2O of positive end-expiratory pressure (oxygen challenge), and mode of donor death. Race was categorized as white, African American, or other. Smoking history was a dichotomous variable, where "yes" indicated any tobacco exposure and "no" represented lifelong nonsmoker. Oxygen challenge was treated as a continuous variable and with a cutoff chosen at 400 mm Hg. Mode of death was categorized as head trauma, cerebrovascular accident, or other. Donor age was assessed as a continuous variable, and then broken down into 5-year increments.
Recipient and surgical variables tested for association with PGF included the following: recipient gender; preoperative diagnosis; recipient age; race; procedure type (single or bilateral); use and duration of cardiopulmonary bypass; pulmonary arterial systolic pressure (PASP), pulmonary arterial diastolic pressure (PADP), and right atrial pressure (RAP) at the time of transplant; ischemic time; use and dose of inhaled nitric oxide; and lowest intraoperative BP. Preoperative diagnosis contains five categories: (1) COPD, including emphysema, chronic bronchitis, bronchiectasis, and bronchiolitis; (2) primary pulmonary hypertension (PPH), exclusive of secondary to congenital cardiac or vascular abnormalities; and (3) cystic fibrosis (CF); (4) fibrotic lung diseases including sarcoidosis and pulmonary fibrosis; and (5) "other," including eosinophilic granuloma, pulmonary alveolar microlithiasis, lymphangioleimatosis, and causes of secondary pulmonary hypertension.
The two major types of lung transplantation procedures are single and bilateral lung transplants. Ischemic time represents the time period that the lung is excluded from circulation, beginning with donor aortic cross-clamp and initiation of pulmonary artery flush and ending with reperfusion of donor lung after completion of all anasthomoses.
Pulmonary artery pressures and RAPs were measured using a pulmonary artery catheter at the time of transplant at the initial reading prior to anesthesia induction. Intraoperative BP was assessed in three ways: first, the lowest systolic BP recording was assessed as a continuous variable; second, it was broken into quartiles; and third, cutoffs were chosen at 70 mm Hg, 65 mm Hg, and 60 mm Hg and analyzed as categorical variables. Use of cardiopulmonary bypass was analyzed both as a dichotomous "yes/no" variable and according to time on bypass. Use of inhaled nitric oxide during the transplant procedure was treated as a dichotomous yes/no variable and according to dose administered. Nitric oxide use initiated in the time period following the operation was not assessed as a risk factor because it was used postoperatively to treat PGF in some cases, and thus may have represented reversal of cause and effect in these cases.
In addition, we chose two variables based on potential confounding effects on our hypothesized risk factor variables. The date of transplantation was chosen (sequential dated number) as a potential confounder based on the proposed "learning curve" of lung transplantation. Likewise, we assessed the surgeon performing the operation as a potential confounder.
The use of antilymphocyte globulin induction therapy was difficult to examine as a risk factor in our population because of the patterns of administration. Sixty-five of the first 100 patients in our cohort study received induction therapy, most with a 3-day course of antilymphocyte immune globulin. Following patient 100, all patients received a 3-day course of induction therapy; therefore, the effects of antilymphocyte therapy could not be accurately evaluated in our study, because the vast majority of our population were treated.
Data Collection
All of the data were collected from review of preexisting medical records. For this study, a specific structured chart abstraction instrument was designed. The data extraction took place in two stages. First, using the instrument, the medical records of all patients in the cohort were reviewed to obtain detailed information on all potential risk factors. This was performed prior to and independent of identification of cases, to minimize information bias on the part of the data collector. Second, following risk factor determination, a structured review of all medical records was performed in order to identify all cases of PGF as previously defined.
Analysis
Univariate Analysis of Candidate Risk Factors and Biomarkers:
Each candidate risk factor was first analyzed as a separate risk factor for the outcome under study by comparing the patients who had PGF with those in the cohort who did not. For discrete variables (such as gender and procedure type), incidences of PGF among those exposed and unexposed were calculated and compared by 
2 or exact tests. RRs, tests of significance, and 95% confidence intervals (CIs) were calculated in the standard fashion.14
The Student t test (for approximately normal data) or the Wilcoxon rank-sum test (for ordinal or markedly nonnormal data for which transformations were either not appropriate or not successful) were used for continuous variables, such as pulmonary artery pressure or age. This univariate analysis was used to select variables for inclusion in the multivariable regression model if they exhibited significance at an 
level of 0.20.15
16
Many of the continuously distributed risk factor variables had the potential for a nonlinear risk for development of PGF. For these variables (such as donor age), data were log transformed and/or broken into categories and assessed for association with PGF. In all analyses containing donor variables, we used a clustering technique to avoid biases introduced by including duplicated information from donors of lungs to two separate patients. In the recipient diagnosis category, sarcoidosis and pulmonary fibrosis were first analyzed separately and then collapsed into a single category due to homogeneity of risk and small numbers in each individual category.
Ischemic time was analyzed in several ways: as a continuous variable, following log transformation, in quartiles; and with cutoffs created at 4 h, 5 h, and 6 h for single lungs and at 6 h, 7 h, and 8 h for bilateral lung transplants (as suggested by prior studies.17 18 ) As a risk factor, it was analyzed separately in single and bilateral transplants. In this analysis, for bilateral sequential lung transplantation, the ischemic time analyzed was the longer ischemic time for the second lung implanted. When ischemic times were used for adjustment in multivariable models, both the first lung and second lung ischemic times were included in tandem in all models. This "nested main effects" method allowed us to account separately for the effects of the first lung ischemic time and the incremental effect of the second lung ischemic time in the entire study population.19
Multivariable Explanatory Analysis: The purpose of the multivariable explanatory analysis was to test the strength and independence of the association of each significant risk factor with development of PGF. Risk factors with a level of significance defined as p < 0.20 in univariate analysis were adjusted for in the multivariable explanatory analysis. In this analysis, each candidate risk factor was evaluated after adjusting for other candidate risk factors and for potential confounding variables, using stratification and Mantel-Haenszel methods for discrete data and logistic regression for continuous data.20 21 To avoid model overfitting, significant risk factors were adjusted for confounding one variable at a time. Confounding was defined as a difference of > 15% between unadjusted and adjusted RRs.22 For a given risk factor variable, all significant confounding variables (ie, those that changed the unadjusted odds ratio [OR] by > 15%) were then included in a multiple logistic regression model. Some potential confounders known to be clinically important in certain situations, such as PASP in the analysis of PPH, were forced into models even though they did not appear to be statistically significant in univariate analysis. A p < 0.05 following adjustment for all confounding variables was considered significant. Risk factor variables with a significant association with PGF following adjustment for confounding were fitted into a final multiple logistic regression model. In addition, variables were assessed for collinearity and effect modification (interaction), using logistic regression and Mantel-Haenszel methods.20 21
To investigate the potential confounding effects of a learning curve phenomenon, risk factors were also adjusted for sequential transplant number as a continuous variable. To investigate potential changes in the incidence of PGF over time, we broke the population into sequential quintiles and performed a 2 test for trend across the quintiles.
Missing Data: Data on candidate risk factors related to recipient and surgical characteristics and patient outcomes were > 95% complete; however, there were some missing data among donor variables. In instances where the donor variable had missing data exceeding 5%, RRs for other significant variables were tested in the data set with missing variables and compared with the complete data set. For all variables, the univariate RR for associations of other risk factor variables (such as PPH or donor age) differed by not more than 10% from the complete data set. The largest amount of missing data occurred in the donor mode of death variable (197 patients available, 78% complete).
This research protocol was approved by the Institutional Review Board of the Office of Regulatory Affairs at the University of Pennsylvania. All statistical comparisons were performed using STATA version 7.0 (STATA Corporation; College Station, TX).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2 test for trend statistic = 3.13, p = 0.54). The characteristics of the individual patients with PGF are listed in Table 1
.
|
|
|
Multivariable Explanatory Analysis of Candidate Risk Factors for Development of PGF
Based on the univariate analysis, the variables eligible to be treated as risk factors in the multivariate analysis were as follows: recipient diagnosis of PPH, recipient gender, recipient age, PASP, ischemic times for both single and bilateral lungs, donor African-American race, donor gender, and donor age. These variables were adjusted for all of the other risk factor variables and potential confounder variables to examine the robustness and independence of the relationship with PGF and assess for the presence of confounding and/or interaction.
There was no confounding of the relationship between PPH and PGF by the following notable variables: use of cardiopulmonary bypass, ischemic time, recipient gender, or any donor variables. In all cases, the relationship of PPH remained associated with PGF (p < 0.02) when adjusted for variables individually. Confounding was detected only by PASP, which increased the RR to 9.24 (95% CI, 1.75 to 48.8) following adjustment (p = 0.009). In addition, there was no interaction detected between PPH and transplant PASP, recipient gender, or transplant type.
Donor age remained a robust risk factor for PGF when adjusted for recipient age, donor gender and race, recipient diagnosis of PPH, donor mode of death, and all other variables. Likewise, donor gender remained significantly associated with PGF when adjusted for all other variables. Donor African-American race became significantly associated with PGF when adjusted for donor gender and donor age. These findings indicate that for a given donor gender or age group, African-American donors had increased risk of acquiring PGF. There was no change in RR when adjusted for all other recipient variables, including recipient race.
Recipient gender lost statistically significant relationship with PGF when adjusted for donor gender. There was no interaction detected between these two variables. PASP had significant confounding when adjusted for the diagnosis of PPH, indicating that PPH was accounting for the majority of the association seen in the univariate analysis. Recipient age, gender, and ischemic time had no significant association with PGF in the multivariable analysis.
To investigate the effect of gender and race mismatching on risk of PGF, two-by-two interactions were tested between recipient and donor gender, and recipient and donor race, and were not significant. Individual combinations of donor and recipient race, and donor and recipient gender are represented in Tables 4 , 5 .
|
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ischemia reperfusion injury following lung transplantation is a spectrum. We chose to define the most severe cases, most similar to ARDS. An important finding in this study is the association of our clinical definition with pathologic findings of diffuse alveolar damage, indicating that our patients had both a severe clinical syndrome and pathology consistent with ischemia-reperfusion injury as the cause. Comparison with other studies of PGF or early graft failure may be hampered by heterogeneity of definition of outcome.5 11 13 More liberal definitions of PaO2/FIO2 ratio, such as those used by Thabut and colleagues,13 may be associated with less severe outcomes, and may be more susceptible to outcome misclassification by including patients with milder forms of lung injury. Definitions based solely on pathology, such as those used by Fisher and colleagues,23 although potentially useful for studying relationships with some outcomes (such as bronchiolitis obliterans), may not have relevance to the full clinical syndrome of PGF that we describe. Similar to the slightly varied definitions of ARDS and acute lung injury,12 different definitions of PGF may be useful for different research and clinical purposes.
The relationship of donor African-American race with PGF appears to be related to the race of the donor, and not to an effect of race-mismatched donors and recipients; however, the number of patients is too small to know this with certainty, particularly in the case of African-American donors paired with African-American recipients. Importantly, the effect of donor race on PGF risk was not related to other factors specific to donors, such as age, mode of death, smoking history, and oxygen challenge. Supporting our findings, Keck et al24 found a significant association between 1-year mortality and the use of lung allografts derived from African-American donors in a multivariate analysis that included many of the variables we analyzed. Similarly, our results are consistent with a previous study25 revealing significantly lower 1-year kidney allograft survival with use of African-American donors, independent of human leukocyte antigen matching. Mechanisms for the observed effect of donor race remain unknown but may reflect differences in vascular endothelium (such as expression of angiotensin-converting enzyme),26 27 28 29 30 31 which could potentially predispose African Americans to more severe ischemia reperfusion injury.
Similar to donor race, the association between donor gender and risk of PGF seems to be most clearly associated with the gender of the donor, rather than other donor factors or with gender mismatching. Gender mismatching is reported as a risk factor for 1-year mortality in the Nineteenth Official Report of the Registry of the International Society for Heart and Lung Transplantation (ISHLT), with an OR of 1.19 (95% CI, 1.03 to 1.37).32 In contrast, our study had the highest rate of PGF among female donors and female recipients (23%) and among female donors and male recipients (20%), with little evidence of gender mismatching driving the relationship. Female gender has been associated with a higher risk of the development of ARDS in the Ibuprofen in Sepsis Study Group,33 as well as in a cohort study of trauma patients.34 Possible mechanisms for these findings are unclear. In other organs, authors have reported inferior posttransplant outcomes with use of female donors following liver35 36 and kidney transplantation.37
The observation of an increased risk of PGF among African-American and female donors in our population should prompt further investigation into the underlying mechanisms and, ideally, into pretransplant therapies aimed at minimizing this risk when these donor populations are utilized. Until further corroborated with other studies, our results should be interpreted with caution and do not justify changes in donor-recipient matching practices at this time.
The finding of PPH as a risk factor for PGF is supported by several prior studies. Other authors4 11 38 have reported a significant association of pulmonary hypertension with serious reperfusion injury in smaller numbers of patients. Likewise, in a study utilizing the United Network for Organ Sharing/ISHLT, Keck et al24 reported a significant early mortality among patients with PPH, possibly due to this increased incidence of PGF and the ensuing increased mortality. Further, in the Nineteenth Annual ISHLT report, PPH was associated with an OR of 1.47 for 1-year mortality (95% CI, 1.08 to 2.02).32 In our study, this relationship was independent of ischemic time, type of transplant procedure, and measured hemodynamics including PASP, PADP, and RAP. There may be important hemodynamic effects (such as change in pressure over time during reperfusion) that may not have been captured by our right-heart pressure measurements.38 Aside from hemodynamic causes, potential explanations for the association between PPH and PGF include use of blood products during the operation for reversal of preoperative anticoagulation, or an intrinsic inability to handle the stress of lung transplantation perhaps related to the underlying pathophysiology of PPH.39 40 41
We observed a significant association of donor age with development of PGF. This finding was independent of other notable variables, including mode of death, race, gender, and recipient age. At first glance, the finding that older donors are more likely to have PGF may indicate that the aging lung is more susceptible to ischemia and reperfusion injury, possibly due to changes to the cytokine milieu of lungs from older donors with ischemia5 ; however, in our population, we also observed that the youngest donors (< 21 years old) were also at higher risk. In a multivariate analysis of risk factors for early mortality using the United Network for Organ Sharing/ISHLT registry, Keck et al24 revealed a similar "j-shaped distribution" of risk for 30-day mortality with donor age. In their study, a quadratic transformation of donor age was used, and a similar high risk was seen among the youngest and oldest donors. This relationship with early mortality was present in the most recent ISHLT registry report, with the lowest risk donor age category at age 35 years.32 This relationship between donor age and outcome has also been illustrated in kidney transplants. Cicciarelli et al25 found that donor age ranges from 1 to 10 years and 70 to 90 years had the worst 1-year graft survival rates. In their study, 15- to 40-year-old donors had the best outcomes.
There were several variables of interest that did not have a significant relationship with PGF in our study. Ischemic time was evaluated as linear variable, after log transformation, and with multiple cutoffs chosen. Although the relationship approached statistical significance in the univariate analysis, when adjusted for other variables (such as diagnosis of PPH), the relationship weakened considerably. Meyer et al17 reported a relationship with ischemic times > 7 h and age of recipient; however, we found no interaction of ischemic time with donor age. Likewise, Thabut and colleagues42 found a significant association with cold ischemic time and poor outcomes among those patients with PGF. In contrast, other groups43 44 have found no association between ischemic times and graft dysfunction. In our study, we found no significant association with PGF among longer ischemic time categories, or with log transformations of the variable, nor did we find any significant interaction with risk factor variables. These findings may indicate that, for most patients in our study, ischemic times were possibly below a threshold level for significant contribution to ischemia reperfusion injury.45
In our study, we included many patients with high PASPs in the non-PPH group, including patients with significant pulmonary hypertension secondary to congenital shunts and chronic pulmonary diseases. PASP and other hemodynamic measures had significantly weakened associations with PGF when adjusted for recipient diagnosis of PPH. These findings indicate that the diagnosis of PPH was accounting for much of the association of elevated pulmonary arterial pressures with PGF. Among other diagnosis subgroups, there was no statistically significant association of pulmonary artery pressures with PGF.
Our study has several limitations. First, it is a single center and may have limited generalizability to other transplant centers. In particular, more variability in patient populations and especially operative practice could potentially yield different risk factors. Second, our incidence of PGF was 11.8%, which may represent an incidence (ie, > 10%) such that ORs obtained from logistic regression analysis might be an overestimate of true RR. Third, although our significant risk factors were adjusted for the confounding effects of other variables, the potential for uncontrolled confounding exists. Potential confounders that we were unable to account for in the present study design include the exact quantity of blood products during operation, and other donor factors, such as exact modes and duration of donor mechanical ventilation prior to transplant. Fourth, although we rigorously defined our exposures and outcomes, the potential for bias due to misclassification exists in our study. For example, our measurement of intraoperative hypotension may not have fully accounted for the duration of hypotension. Outcome misclassification seems less likely in our population, given our definition of PGF and exclusion of other causes of similar hypoxic syndromes (eg, pulmonary venous obstruction). Any misclassification would likely be nondifferential and thus would bias the results toward the null. Fifth, given the multiple risk factors examined in our study, the possibility for type I error exists. We tried to limit this by only testing risk factors with a biologically plausible association with PGF. In addition, to avoid overfitting the models, we limited the number of variables analyzed in a given model such that there were at least 5 to 10 outcomes per covariate. Finally, there may have been some modest effects that we were not able to detect due to type II error. Although we had reasonable power to detect individual risk factors for PGF, some of our assessments of interaction between variables were limited by our sample size. Thus, we were unable to fully assess interaction between important risk factors.
In the present study, we have identified several important risk factors for the development of PGF. In general, the prominence of demographic variables specific to donors and recipients rather than operative variables may indicate the importance of factors inherent to individual donors and recipients that may shape the response to ischemia and reperfusion. Confirmation of these findings and further research into the underlying mechanisms responsible for these associations may lead to selective therapies aimed at preventing this devastating posttransplant lung injury syndrome.
|
Footnotes |
|---|
Presented in part at CHEST 2001, November, 2001, Philadelphia, PA; and at the American Thoracic Society International Conference, May 2001, San Francisco, CA.
Supported by NHLBI K23, HL04243, and the Craig and Elaine Dobbin Fund.
Received for publication January 8, 2003. Accepted for publication March 31, 2003.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Oto, A. P. Griffiths, F. Rosenfeldt, B. J. Levvey, T. J. Williams, and G. I. Snell Early Outcomes Comparing Perfadex, Euro-Collins, and Papworth Solutions in Lung Transplantation Ann. Thorac. Surg., November 1, 2006; 82(5): 1842 - 1848. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Oto, A. P. Griffiths, B. J. Levvey, D. V. Pilcher, T. J. Williams, and G. I. Snell Definitions of primary graft dysfunction after lung transplantation: Differences between bilateral and single lung transplantation J. Thorac. Cardiovasc. Surg., July 1, 2006; 132(1): 140 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Mathur, M. Baz, E. D. Staples, M. Bonnell, J. M. Speckman, P. J. Hess Jr, C. T. Klodell, D. G. Knauf, L. L. Moldawer, and T. M. Beaver Cytokine profile after lung transplantation: correlation with allograft injury. Ann. Thorac. Surg., May 1, 2006; 81(5): 1844 - 1850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Whitson, D. S. Nath, A. C. Johnson, A. R. Walker, M. E. Prekker, D. M. Radosevich, C. S. Herrington, and P. S. Dahlberg Risk factors for primary graft dysfunction after lung transplantation J. Thorac. Cardiovasc. Surg., January 1, 2006; 131(1): 73 - 80. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Hartwig, J. Z. Appel III, E. Cantu III, S. Simsir, S. S. Lin, C.-C. Hsieh, R. Walczak, S. M. Palmer, and R. D. Davis Jr Improved Results Treating Lung Allograft Failure With Venovenous Extracorporeal Membrane Oxygenation Ann. Thorac. Surg., November 1, 2005; 80(5): 1872 - 1880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Oto, M. Rabinov, A. P. Griffiths, H. Whitford, B. J. Levvey, D. S. Esmore, T. J. Williams, and G. I. Snell Unexpected donor pulmonary embolism affects early outcomes after lung transplantation: A major mechanism of primary graft failure? J. Thorac. Cardiovasc. Surg., November 1, 2005; 130(5): 1446 - 1446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Sandmeier, R. Speich, E. Grebski, P. Vogt, E. W. Russi, W. Weder, and A. Boehler Iron Accumulation in Lung Allografts Is Associated With Acute Rejection But Not With Adverse Outcome Chest, September 1, 2005; 128(3): 1379 - 1384. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Rubin and D. B. Badesch Evaluation and Management of the Patient with Pulmonary Arterial Hypertension Ann Intern Med, August 16, 2005; 143(4): 282 - 292. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Oto, B. J. Levvey, D. V. Pilcher, M. J. Bailey, and G. I. Snell Evaluation of the oxygenation ratio in the definition of early graft dysfunction after lung transplantation J. Thorac. Cardiovasc. Surg., July 1, 2005; 130(1): 180 - 186. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Christie, R. M. Kotloff, V. N. Ahya, G. Tino, A. Pochettino, C. Gaughan, E. DeMissie, and S. E. Kimmel The Effect of Primary Graft Dysfunction on Survival after Lung Transplantation Am. J. Respir. Crit. Care Med., June 1, 2005; 171(11): 1312 - 1316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D.V. Pilcher, C.D. Scheinkestel, G.I. Snell, A. Davey-Quinn, M.J. Bailey, and T.J. Williams High central venous pressure is associated with prolonged mechanical ventilation and increased mortality after lung transplantation J. Thorac. Cardiovasc. Surg., April 1, 2005; 129(4): 912 - 918. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Christie Lung Allograft Ischemic Time: Crossing the Threshold Am. J. Respir. Crit. Care Med., April 1, 2005; 171(7): 673 - 674. [Full Text] [PDF]
|
|
|
|
|
|
G. Thabut, H. Mal, J. Cerrina, P. Dartevelle, C. Dromer, J.-F. Velly, M. Stern, P. Loirat, G. Leseche, M. Bertocchi, et al. Graft Ischemic Time and Outcome of Lung Transplantation: A Multicenter Analysis Am. J. Respir. Crit. Care Med., April 1, 2005; 171(7): 786 - 791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Christie, J. S. Sager, S. E. Kimmel, V. N. Ahya, C. Gaughan, N. P. Blumenthal, and R. M. Kotloff Impact of Primary Graft Failure on Outcomes Following Lung Transplantation Chest, January 1, 2005; 127(1): 161 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. B. Marshall and L. J. Kohman Gender and lung transplantation: Size matters, does sex? J. Thorac. Cardiovasc. Surg., September 1, 2004; 128(3): 352 - 353. [Full Text] [PDF]
|
|
|
|
|
|
