| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 2005;130:180-186
© 2005 The American Association for Thoracic Surgery
Cardiothoracic Transplantation |
a Department of Cardiothoracic Surgery, Allergy, Immunology, The Alfred Hospital, Monash University, Melbourne, Victoria, Australia.
b Department of Respiratory Medicine, The Alfred Hospital, Monash University, Melbourne, Victoria, Australia.
c Department of Intensive Care Medicine, The Alfred Hospital, Monash University, Melbourne, Victoria, Australia.
d Department of Epidemiology and Preventive Medicine, The Alfred Hospital, Monash University, Melbourne, Victoria, Australia.
Received for publication August 4, 2004; revisions received October 13, 2004; accepted for publication October 28, 2004. * Address for reprints: Takahiro Oto, MD, Department of Cardiothoracic Surgery, The Alfred Hospital, Commercial Road, Melbourne, VIC 3004, Australia (Email: takahirooto{at}aol.com).
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionReferences |
|---|
METHODS: A retrospective study was performed that included all 68 bilateral single-lung transplantations at The Alfred Hospital from January 2000 to December 2002.
RESULTS: PaO 2/FIO 2 at 6 and 12 hours after admission to the intensive care unit was significantly associated with the duration of intubation (r = –0.44; P < .001 and r = –0.48; P < .001, respectively), and PaO 2/FIO 2 at 6 and 24 hours was also significantly associated with the length of intensive care unit stay (r = –0.38; P = .002 and r = –0.44; P = .001, respectively). Thirty-day mortality was significantly associated with a lower PaO 2/FIO 2 at 6 hours (219 ± 93 vs 306 ± 101; P = .03).
CONCLUSIONS: PaO 2/FIO 2 taken between 6 and 12 hours after transplantation is a useful marker associated with lung transplantation outcomes. There is the potential for therapeutic interventions during this time that may be able to enhance PaO 2/FIO 2 by 12 hours and improve subsequent outcomes.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
|
Lung transplantation has become an accepted therapeutic option for end-stage pulmonary disease and has resulted in remarkable overall improvements in outcome.
1
However, despite advances in organ preservation, surgical techniques, and perioperative care, early mortality after lung transplantation remains notable. According to the 2003 Registry of the International Society for Heart and Lung Transplantation, 41% of 1-year deaths occur within 30 days of transplantation. Early graft dysfunction is one of the leading causes of this 30-day mortality, in turn, accounting for 30.5% of deaths in this time period.
2
Early graft dysfunction encompasses other terms, such as primary graft failure, pulmonary reimplantation response, and reperfusion edema, and is clinically recognized as a impaired gas exchange, opacification of the chest radiograph, and increased pulmonary vascular resistance within 24 hours of transplantation.
3–7
Whatever the original stimulus, this syndrome results in the requirement for prolonged mechanical ventilatory support with high levels of inspired oxygen and a prolonged stay in the intensive care unit (ICU).
3–9
Despite the clinical importance of early graft dysfunction, no standardized definition or scoring system is available to assess its severity and to predict its outcome.
3–9
A standardized definition of early graft dysfunction enables us to grade the severity of the graft function and also to create a standardized therapeutic strategy for early graft dysfunction that is relevant to reducing early morbidity and mortality and their associated costs. There are also advantages for the evaluation of management strategies for donors and recipients and comparison between centers.
The arterial blood gas oxygen tension/inspired oxygen fraction ratio (PaO
2/FIO
2) is a simple marker widely used to assess the graft function in lung transplantation
3–9
and to define adult respiratory distress syndrome.
10
We hypothesized that PaO
2/FIO
2 would prove to be a useful marker associated with subsequent early outcomes in the definition of early graft dysfunction after lung transplantation. The aims of this study were to define the prevalence of various ranges of PaO
2/FIO
2 over the first 48 hours after lung transplantation and to evaluate which measurement using PaO
2/FIO
2 best correlates with the duration of intubation, the length of the ICU stay, and 30-day survival, which are important alternative indicators of early graft performance.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Transplantation Protocol
Donor assessment, recipient selection, donor/recipient matching, surgical technique, and postoperative management proceeded according to our standard protocol, which has been described elsewhere.
11–14
Although lung donor selection criteria are based on standard criteria, extended donors are commonly considered at our institution.
15
In some of these cases, the donor chest radiograph may be abnormal, secretions may be purulent on bronchoscopy, and the donor PaO
2/FIO
2 may be less than 300 mm Hg.
Preoperative immunologic evaluation was routinely performed. The presence of preformed antibodies to human leukocyte antigen was screened with a panel-reactive antibody assay, and all transplant recipients were found to have values less than 10%. A prospective donor/recipient T-cell and B-cell cross-match was performed in all cases.
Lung procurement and preservation followed standard procedures with cold-modified Euro-Collins solution for antegrade flushing. However, we infused prostacyclin (Flolan) at 40 to 80 ng · kg–1 · min–1 intravenously for approximately 10 minutes before crossclamp.
Postoperative management in the ICU was performed to ensure satisfactory end-organ perfusion while maintaining a relatively low filling pressure (cardiac index, >2.4; pulmonary capillary wedge pressure, <10 mm Hg; and central venous pressure, <7mm Hg). Patients with severe early graft dysfunction received a standardized evaluation and therapy with increasing complexity depending on the degree of ventilatory and hemodynamic compromise. Transesophageal echocardiography was performed to exclude lung torsion and pulmonary vascular obstruction, and a retrospective cross-match was performed to exclude humoral rejection. Therapy included pressure-controlled mechanical ventilation, limitation of positive end-expiratory pressure to 12 to 15 cm H2O, negative fluid balance with furosemide or continuous venovenous hemofiltration, inhaled nitric oxide 5 to 20 ppm, and elevation of the upper body or lateral positioning if appropriate. If these approaches failed, then extracorporeal membrane oxygenation was considered.
14
Immunosuppression was based on triple therapy with cyclosporine (trough levels of 300–450 µg/L), azathioprine (1.5–2.0 mg · kg–1 · d–1), and prednisolone (0.15 mg · kg–1 · d–1). Prophylaxis for Pneumocystis carinii and cytomegalovirus infection was achieved with low-dose oral trimethaprim-sulfamethoxazole and intravenous ganciclovir, respectively.
Data Collection
All of the recipient data were collected from review of medical records. Best and worst PaO
2/FIO
2 within 24 and 48 hours and PaO
2/FIO
2 at 0, 6, 12, 18, 24, 36, and 48 hours after admission to the ICU were collected. When blood gas analysis data at a specific time point were not available, then the data closest to the time point were substituted. The donor PaO
2/FIO
2 was the last arterial blood gas result on FIO
2 = 1.0 recorded at the time of donor referral to the transplant center. The graft ischemic time of a bilateral single-lung transplantation was defined as the ischemic time for the second transplanted lungs in this study.
Assessment of Outcome of Early Graft Dysfunction
Duration of intubation, length of ICU stay, and 30-day mortality were used as outcome indicators of early graft dysfunction.
Statistical Analysis
Analysis was performed with SAS version 8.2 (SAS Institute, Inc, Cary, NC). Data were initially assessed for normality and were log-transformed where appropriate. Duration of intubation and length of ICU stay were both well approximated by a log-normal distribution. Univariate analysis was performed on log-transformed intubation time and log-transformed ICU stay by using Pearson correlation coefficients for continuous variables and Student t tests for binomial variables. Multivariate analysis for these 2 outcomes was performed by using multiple linear regression. Univariate analysis for 30-day mortality was performed by using 
2 tests for equal proportion and logistic regression. Multivariate analysis for 30-day mortality was conducted with multiple linear regression. Time to death was further assessed by using Cox proportional hazard regression for continuous variables and log-rank tests for categorical variables. Multivariate models were constructed by using a stepwise selection technique that considered all possible prediction variables. Multivariate models were then validated with a backward-elimination technique and further assessed for biological and clinical plausibility. Continuous data were reported as mean ± standard deviation, whereas categorical data were reported as count and proportions.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Donor and Recipient Demographics
The demographics of donors and recipients are summarized in Table 1. Cerebrovascular accident (51%) was the most common donor cause of death, followed by trauma (34%). Ten percent of the donors had a smoking history of more than 20 pack-years. With 1 exception, all donors had a PaO
2/FIO
2 of more than 300 (mean, 474 ± 95). Most recipients had cystic fibrosis (60%), and the percentage of interstitial lung disease and pulmonary hypertension was 10% and 9%, respectively. Ten percent of the recipients had cardiopulmonary bypass during transplantation.
|
|
|
|
|
|
Multivariate Regression Analysis
In the multivariate analysis, PaO
2/FIO
2 at 6 and 12 hours was the strongest predictor of duration of intubation, and PaO
2/FIO
2 at 6 and 24 hours was also the strongest predictor for length of ICU stay. All of these remained statistically significant after adjusting for any potential confounding effects.
Population Kaplan-Meier Analysis
The percentage of patients not extubated versus time and the percentage of patients staying in the ICU versus time subgrouped by the degree of PaO
2/FIO
2 at 6 hours were estimated by the Kaplan-Meier method. There were significant differences between the curves of subgroups in both intubation time and ICU stay (Figure 4). Fifty-four percent of the patients with PaO
2/FIO
2 at 6 hours less than 200 remained intubated 48 hours after admission to the ICU and also remained in the ICU 7 days after transplantation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
Unfortunately, comparison with and between these other studies of early graft dysfunction is difficult because of the lack of a homogeneous definition. In this study, the continuous variables of duration of intubation and length of ICU stay and the categorical variable of 30-day mortality were used as alternative outcomes of early graft dysfunction to allow for comparison with all prior definitions. PaO
2/FIO
2, a simple parameter without any requirement for complicated methods in assessment and calculation, has been widely used for assessing the graft function in lung transplantation
3–9
and for defining adult respiratory distress syndrome.
10
Results generally match those of this study: Thabut and colleagues
5
reported that the worst PaO
2/FIO
2 within the first 3 postoperative days was associated with ICU mortality, and Chatila and colleagues
8
reported that the best PaO
2/FIO
2 in the first 24 hours was associated with prolonged mechanical ventilation. Lee and colleagues
9
reported that an immediate postoperative PaO
2/FIO
2 of <200 correlated with an ICU stay longer than 5 days. Christie and colleagues
4
included PaO
2/FIO
2 <200 persisting beyond the first 48 hours in the criteria for early graft dysfunction and found it to be associated with prolonged mechanical ventilation and hospital stay. Although PaO 2/FIO 2 has been clarified to be a predictor of early graft dysfunction, no information is available regarding which measurement using PaO 2/FIO 2 best correlates with early graft dysfunction. In this study, PaO 2/FIO 2 at 6 and 12 hours after admission to the ICU was independently and strongly associated with duration of intubation, and PaO 2/FIO 2 at 6 and 24 hours was also independently and strongly associated with the length of ICU stay. A PaO 2/FIO 2 <200 at 6 hours seemed to indicate a prolonged duration of intubation and ICU stay (Figure 3). The PaO 2/FIO 2 at 6 hours was not significant in a final multivariate regression model for analysis of 30-day mortality. It was a significant predictor in a univariate model; however, the number of actual deaths was small.
In a clinical situation, the availability of the markers using PaO 2/FIO 2 at an earlier time point (such as 6 hours) contributes to its usefulness beyond what is revealed by the data at 12 and 24 hours, because this has the potential to allow for earlier intervention for a specific group of patients with early graft dysfunction. Therefore, PaO 2/FIO 2 at 6 hours might be a more useful marker than that at 12 or 24 hours.
The last donor PaO 2/FIO 2 before procurement was also a significant factor associated with the duration of intubation (Table 2). When we looked at the correlation between donor and recipient PaO 2/FIO 2, donor PaO 2/FIO 2 best correlated with recipient PaO 2/FIO 2 at 0 hours (r = 0.54; P < .001). Although donor PaO 2/FIO 2 was associated with the graft oxygenation immediately after transplantation and there was a general trend toward improved graft function during the initial postoperative period up to 12 hours (Figure 1), exactly what transpires in that first 12 hours is still critical because PaO 2/FIO 2 at 12 hours proved to best correlate with the duration of intubation.
Some possible limitations to this study should be considered. First, ventilator settings potentially influenced PaO
2/FIO
2; however, postoperative ventilatory management was performed according to standard protocols, which generally aim toward the highest PaO
2/FIO
2 for minimum ventilator settings. An oxygen index [(mean airway pressure x percentage of inspired oxygen)/partial pressure of arterial oxygen] was reported to be useful for predicting early outcomes after lung transplantation
16
and to be worth further investigation. However, mean airway pressure is not necessarily and easily calculated by all ventilators, is not available on extubated patients, and would not be applicable to single-lung transplantations (not addressed to this study). Second, most patients did not have a complicated postoperative course and were typically extubated on the second postoperative day. Therefore, a spectrum of severity of early graft dysfunction, ranging from a mild form that was represented only by subtle graft edema on chest radiograph to a severe form that required prolonged mechanical ventilatory support, was included in this study. It may be relevant to define only the more severe forms of early graft dysfunction. Third, we combined all causes of early graft dysfunction; however, contusion, clot, and cardiac dysfunction should possibly be removed to distinguish pure primary graft failure.
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Currey, D. V. Pilcher, A. Davies, C. Scheinkestel, M. Botti, M. Bailey, and G. Snell Implementation of a management guideline aimed at minimizing the severity of primary graft dysfunction after lung transplant J. Thorac. Cardiovasc. Surg., January 1, 2010; 139(1): 154 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. McIlroy, D. V. Pilcher, and G. I. Snell Does anaesthetic management affect early outcomes after lung transplant? An exploratory analysis Br. J. Anaesth., April 1, 2009; 102(4): 506 - 514. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Prekker, C. S. Herrington, M. I. Hertz, D. M. Radosevich, and P. S. Dahlberg Early Trends in PaO2/Fraction of Inspired Oxygen Ratio Predict Outcome in Lung Transplant Recipients With Severe Primary Graft Dysfunction Chest, September 1, 2007; 132(3): 991 - 997. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Oto and G. I. Snell Reply to the Editor J. Thorac. Cardiovasc. Surg., July 1, 2007; 134(1): 270 - 271. [Full Text] [PDF]
|
|
|
|
|
|
D. V. Pilcher, G. M. Auzinger, B. Mitra, D. V. Tuxen, R. F. Salamonsen, A. R. Davies, T. J. Williams, and G. I. Snell Predictors of independent lung ventilation: An analysis of 170 single-lung transplantations J. Thorac. Cardiovasc. Surg., April 1, 2007; 133(4): 1071 - 1077. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Oto, B. J. Levvey, H. Whitford, A. P. Griffiths, T. Kotsimbos, T. J. Williams, and G. I. Snell Feasibility and Utility of a Lung Donor Score: Correlation With Early Post-Transplant Outcomes Ann. Thorac. Surg., January 1, 2007; 83(1): 257 - 263. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Oto, A. P. Griffiths, B. J. Levvey, T. J. Williams, and G. I. Snell Unilateral radiographic abnormalities after bilateral lung transplantation: Exclusion from the definition of primary graft dysfunction? J. Thorac. Cardiovasc. Surg., December 1, 2006; 132(6): 1441 - 1446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
T. Oto, M. Rabinov, J. Negri, S. Marasco, M. Rowland, A. Pick, G. Snell, F. Rosenfeldt, and D. Esmore Techniques of Reconstruction for Inadequate Donor Left Atrial Cuff in Lung Transplantation Ann. Thorac. Surg., April 1, 2006; 81(4): 1199 - 1204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Estenne and R. M. Kotloff Update in transplantation 2005. Am. J. Respir. Crit. Care Med., March 15, 2006; 173(6): 593 - 598. [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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
