- Email: [email protected]
Clinical Factors Implicated in Primary Graft Dysfunction After Heart Transplantation: A Single-center Experience
Clinical Factors Implicated in Primary Graft Dysfunction After Heart Transplantation: A Single-center Experience R.A. Quintana-Quezadaa,*, I. Rajapreyara, A. Postalian-Yrausquina, Y.C. Yeha, S. Choib, B. Akkantic, A. Siegd, P. Weeksd, M. Patela, J. Patela, S. Nathana, B. Kara, P. Loyalkaa, and I. Gregorica a
Center for Advanced Heart Failure, University of Texas Health Science Center at Houston/Memorial Hermann Hospital, Texas Medical Center, bDivision of Clinical and Translational Sciences, Department of Internal Medicine, cDivision of Critical Care, Pulmonary and Sleep, University of Texas Medical School at Houston, and dDepartment of Pharmacy, Memorial Hermann e Texas Medical Center, Houston, Texas
ABSTRACT Background. Primary graft dysfunction (PGD) is a frequent complication after cardiac transplantation and remains one of the leading causes of mortality in these patients. The objective of this case-control study is to identify donor and surgical procedure’s factors associated with PGD, and further guide possible strategies to prevent PGD. Methods. Retrospective analysis of the medical records of patients who underwent cardiac transplantation at Memorial Hermann Hospital at Texas Medical Center between October 2012 and February 2015. Results. The study population included 99 patients, of which 18 developed PGD. Univariate analysis of donor characteristics revealed opioid use (P ¼ .049) and death owing to anoxia (P ¼ .021) were associated with PGD. The recipient/donor blood type match AB/A was significantly associated with PGD (P ¼ .031). Time from brain death to aortic cross clamp (TBDACC) of 3 and 5 days were also found to be associated with PGD (P ¼ .0011 and .0003, respectively). Multivariate analysis confirmed that patients with a time from brain death to aortic cross clamp 3 and 5 days had lesser odds of developing PGD (odds ratio, 0.098 [P ¼ .0026] and OR, 0.092 [P ¼ .0017], respectively]. Conclusions. Our study showed that a longer time from brain death to aortic cross clamp was associated with lower odds of developing PGD. Therefore, postponing heart procurement for a few days after brain death seems to be beneficial in preventing PGD.
S
INCE the first cardiac transplantation in 1967, patient outcomes have improved significantly as a consequence of new surgical techniques, more effective immunosuppressive agents, better organ preservation, prophylaxis against opportunistic infections, and early rejection recognition with endomyocardial biopsies [1e5]. Nonetheless, primary graft dysfunction (PGD) remains a leading cause of 30-day mortality in these patients. The true incidence of PGD is hard to estimate owing to reporting variability by different heart transplant centers and the previous lack of a standardized definition. In April 2013, a consensus conference on PGD developed guidelines to define, diagnose, and treat patients with PGD after cardiac transplantation. This consensus included data from 47 transplant centers, comprising a
total of 9901 patients who underwent heart transplants. Among those patients, 733 (7.4%) developed PGD. Of those, 30-day and 1-year mortality was 30% and 35%, respectively. The leading reported causes of death were multiorgan system failure (70%), primary graft failure (20%), and sepsis (10%). PGD was defined as left ventricle (LV), biventricular (BV), or right ventricle (RV) failure that occurs within 24 hours after cardiac transplantation, not attributable to a discernible etiology such
*Address correspondence to Raymundo Alain QuintanaQuezada, MD, University of Texas Health Science Center at Houston, 6410 Fannin St, Suite 920, Houston, Texas 77030. E-mail: [email protected]
0041-1345/16 http://dx.doi.org/10.1016/j.transproceed.2016.02.073
ª 2016 Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
2168
Transplantation Proceedings, 48, 2168e2171 (2016)
PRIMARY GRAFT DYSFUNCTION
as hyperacute rejection, pulmonary hypertension, or a surgical complication, events that would be considered secondary graft dysfunction [6]. The increasing number of patients active on the waiting list for cardiac transplantation, the potential complications that ventricular support devices may be associated with, and the currently limited donor availability may cause the donor selection process to become more lenient. The objective of this case-control study is to identify donor and surgical procedural factors associated with PGD and to further guide possible strategies to prevent PGD.
2169
Statistical Techniques Continuous variables were reported as mean values standard deviation or median values (interquartile range) as appropriate, and compared using the Student t test or the ManneWhitney U test. To compare categorical variables, the c2 test was used or Fisher exact test if expected frequencies were <5. All variables determined to be clinically important or suggested by univariate analysis to be significant (P < .05) were included in the multivariate regression. Multivariable logistic regression analysis was performed to assess the simultaneous effect of multiple variables on PGD. The 95% confidence interval (CI) and odds ratio (OR) were reported for each variable. All statistical analyses were performed using SAS software (version 9.3, SAS Institute, Cary NC).
MATERIAL AND METHODS We retrospectively analyzed the medical records of all patients who underwent cardiac transplantation at Memorial Hermann Hospital in the Texas Medical Center between October 2012 and February 2015. PGD was defined as LV, BV, or RV failure that occurred within 24 hours after cardiac transplantation and was not attributable to a surgical complication; this was done according to the definition established by the consensus conference on PGD after cardiac transplantation [6]. Patients with secondary graft dysfunction were excluded; hyperacute rejection was ruled out by performing endomyocardial biopsies, and we did not include patients who developed RV failure owing to elevated pulmonary vascular resistance. Patients with PGD affecting the LV (PGD-LV) were further classified into 3 grades of PGD based on the echocardiographic data, hemodynamic data, and the type of support (pharmacologic or mechanical) required. Mild PGD-LV included patients with an LV ejection fraction of 40% or right atrial pressure of >15 mm Hg, pulmonary capillary wedge pressure of >20 mm Hg, and a cardiac index of <2.0 L/min/m2 (for >1 hour) who required low-dose inotropic support (inotrope score < 10). Moderate PGD-LV included patients with an LV ejection fraction of 40% or right atrial pressure of >15 mm Hg, pulmonary capillary wedge pressure of >20 mm Hg, and cardiac index of <2.0 L/min/m2 (for >1 hour) requiring high-dose inotropic support (inotrope score > 10); and placement of an intraaortic balloon pump. Severe PGD-LV included patients dependent on LV or BV mechanical circulatory support including extracorporeal membrane oxygenation, an LV assist device, BV assist device, or percutaneous LV assist device not including intraaortic balloon pump. PGD-RV is often harder to quantify; thus, no severity grading system was established. The diagnosis of PGD-RV was established in patients with a right atrial pressure of >15 mm Hg, pulmonary capillary wedge pressure of <15 mm Hg, cardiac index of <2.0 L/min/m2, and transpulmonary gradient of <15 mm Hg and/or pulmonary artery systolic pressure of <50 mm Hg or in any patient requiring RV assist device placement [6]. Donor and recipient characteristics were studied in patients who developed PGD and then compared with patients without PGD (control group). Drug dosages were converted to micrograms per kilogram per minute and the inotrope score was calculated according to the following formula: dopamine (dose 1) þ dobutamine (dose 1) þ epinephrine (dose 100) þ norepinephrine (dose 100) þ isoproterenol (dose 100) þ milrinone (dose 15) [7,8]. The doses of vasopressors, phenylephrine and vasopressin were converted to micrograms per kilogram per minute and units per kilogram per minute, respectively, and analyzed separately. Data collection, analysis, and reporting were approved by the institutional review board.
RESULTS
We reviewed a total of 102 charts. Three patients were not included in our study owing to incomplete donor records. We found 18 patients who developed PGD, and 81 served as controls. Among the patients with PGD, 4 developed PGDRV, 9 had severe PGD-LV, 2 had moderate PGD-LV, and 3 had mild PGD-LV. The donor and surgical procedure characteristics are shown in Tables 1 and 2, respectively. Among the donor characteristics, a history of opioid use (P ¼ .049) and death owing to anoxia (P ¼ .021) were associated significantly with PGD. The recipient/donor blood type match AB/A was significantly associated with PGD (P ¼ .031). Time from brain death to aortic cross clamp (TBDACC) of 3 and 5 days were also associated with the development of PGD (P ¼ .0011 and .0003, respectively). Multivariate analysis (Table 3) confirmed that patients with a TBDACC 3 days (OR, 0.098; 95% CI, 0.021e0.443; P ¼ .0026) and 5 days (OR, 0.092; 95% CI, 0.010e0.411; P ¼ .0017) had progressively lower odds of developing PGD. DISCUSSION
In our study, the rate of PGD was 18.18%, which is higher than the reported incidence in the consensus conference on PGD [6]. Upon reviewing the donor characteristics, we were unable to find a difference in the number of blood transfusions, inotrope score, phenylephrine dose, vasopressin dose, preharvest donor down time, and ischemic time between the PGD and control group. These variables are wellknown to be associated with PGD [9e11]. However, we believe our sample size was not large enough to detect a significant difference. Regarding causes of death, despite previous reports describing trauma as an independent risk factor for the occurrence of primary graft failure [10], our study showed an association between anoxia and PGD (P > .021), which disappeared in the multivariate analysis (OR, 4.051; 95% CI, 0.846e19.4; P ¼ .080). Multiple animal and human studies have shown that brain death induces release of catecholamines, cytokines and biologic mediators which, through a series of molecular events, cause deterioration of myocardial contractility and predispose the heart to ischemic injury [12e17]. Cantin et al. [18]
2170
QUINTANA-QUEZADA, RAJAPREYAR, POSTALIAN-YRAUSQUIN ET AL Table 1. Univariate Analysis of Donor’s Clinical Factors Donor Factors
Without PGD, n/N (%)
Gender difference (female)* Age (y) <20* 20e29* 30e39* 40e49* BMI (kg/m2) <20* 20e24* 25e29* 30e34* >35* Diabetes mellitus* Hypertension* CPR* Downtime < 10 min* Downtime >10 min* TBDACC < 5 d* TBDACC > 5 d* TBDACC < 3 d* TBDACC > 3 d* Marijuana use* Benzodiazepine use* Cocaine use* Amphetamine use* Opioid use* Phencyclidine use* Nonsmokers* Smokers (1e10 packs-year)* Smokers (>10 packs-year)* Mean inotropic score† Phenylephrine dose (mg/kg/min), mean SD† Vasopressin dose, mean SD (U/kg/min)† Troponin I < 1* Troponin I > 1* Ejection fraction (>50%)* Prolonged QTc Anoxia* Traumatic brain injury* Intracerebral* hemorrhage Gunshot wound*
21/81 (25.92%) 12/81 35/81 25/81 9/81
(14.81) (43.21) (30.86) (11.11)
13/81 (16.05) 23/81 (28.40) 28/81 (34.57) 12/81 (14.81) 5/81 (6.17) 3/81 (3.70) 10/81 (12.35) 25/81 (30.86) 67/81 (82.72) 14/81 (17.28) 11/81 (13.58) 70/81 (86.42) 28/81 (34.57) 54/81 (66.67) 27/81 (33.33) 11/81 (13.58) 4/81 (4.94) 8/81 (9.88) 8/81 (9.88) 1/81 (1.23) 48/81 (59.26) 25/81 (30.86) 8/81 (9.88) 16.8 0.341 0.583 0.0004 (0.003) 60/81 (74.07) 22/81 (27.16) 68/81 (83.95) 31/81 (38.27) 13/81 (16.05) 23/81 (28.4) 16/81 (19.75) 16/81 (19.75)
PGD, n/N (%)
2/18 (11.11) 2/18 10/18 3/18 3/18
P
.231
(11.11) (55.56) (16.67) (16.67)
.99 .311 .384 .444
1/18 (5.56) 6/18 (33.33) 7/18 (38.89) 3/18 (16.67) 1/18 (5.56) 2/18 (11.11) 3/18 (16.67) 8/18 (44.44) 11/18 (61.11) 7/18 (38.89) 10/18 (55.55) 8/18 (44.44) 14/18 (77.78) 3/18 (16.67) 2/18 (11.11) 3/18 (16.67) 2/18 (11.11) 4/18 (22.22) 5/18 (27.78) 0/18 (0) 11/18 (61.11) 4/18 (22.22) 3/18 (16.67) 11.7 0.335 (0.681) 0.001 (0.004) 10/18 (55.56) 7/18 (38.89) 17/18 (94.44) 9/18 (50) 8/18 (44.44) 2/18 (11.11) 6/18 (33.33) 2/18 (11.11)
.454 .775 .787 .731 .99 .219 .699 .277 .55 .0003 .0011 .574 .156 1 .221 .049 1 1 .577 .411 .254 .990 .338 .384 .294 .285 .021 .227 .217 .515
Abbreviations: BMI, body mass index; CPR, cardiopulmonary resuscitation; TBDACC, time from brain death to aortic cross clamp; QTc, corrected QT interval; SD, standard deviation. *Fisher exact test. † 2 c test.
studied 475 heart transplant cases and found that patients with a shorter management time; defined as time from an incident leading to brain death (ie, trauma, subarachnoid hemorrhage) until aortic cross clamp, had an increased survival when compared with patients with a shorter management time. When they specifically evaluated the TBDACC, they did not find a significant association with survival. The authors concluded that events before the time point when the patient developed brain death somehow damaged the donor heart [18]. In our study, we found that patients with a TBDACC of 3 days had significantly lower odds of developing PGD compared to patients with a TBDACC of <3 days.
When we analyzed the patients with a TBDACC of 5 days we found that their odds of developing PGD were even lower. Similar findings have been described in renal transplant cases, in which a longer TBDACC decreased the risk of delayed graft function [19]. The proposed explanation, based on a few animal studies done in kidney [20] and lung [21] transplant models, is that if more time elapses after brain death, adaptive tissue mechanisms to recover from the initial catecholamine surge injury can take place. The study is limited because it is retrospective and we are not considering other complications of heart transplantation, including hyperacute rejection episodes and surgical complications.
PRIMARY GRAFT DYSFUNCTION
2171
Table 2. Univariate Analysis of Surgical Procedure’s Factors Surgical procedure factors
Without PGD
Gender matches* (recipient-donor) Male to male 42/81 (51.85%) Male to female* 19/81 (23.46%) Female to female* 9/81 (11.11%) Female to male* 13/81 (16.05%) CMV status* R ()/D(þ) 18/81 (22.22%) R()/D() 16/81 (19.75%) R(þ)/D() 12/81 (14.81%) R(þ)/D(þ) 35/81 (43.21%) Ischemic time (h)* <2 13/81 (16.05%) 2e4 59/81 (72.84%) >4 9/81 (11.11%) Blood type matches (R/D)* O/O 31/81 (38.27%) A/O 4/81 (4.94%) B/O 5/81 (6.17%) A/A 28/81 (34.57%) B/B 10/81 (12.35%) AB/AB 2/81 (2.47%) AB/A 0/81 (0%) AB/B 1/81 (1.23%) PRBCs transfusion >10* 12/81 (14.81%)
PGD
P
14/18 2/18 0/18 2/18
(77.78%) (11.11%) (0%) (11.11%)
.065 .348 .357 .999
1/18 4/18 4/18 9/18
(5.56%) (22.22%) (22.22%) (50%)
.0477 .751 .183 .613
2/18 (11.11%) 13/18 (72.22%) 3/18 (16.67%) 6/18 0/18 1/18 6/18 3/18 0/18 2/18 0/18 3/18
(33.33%) (0%) (5.55%) (33.33%) (16.67%) (0%) (11.11%) (0%) (16.67%)
.9
.791 .999 .999 .999 .698 .999 .031 .999 .731
Abbreviations: CMV, cytomegalovirus; D, donor; PGD, primary graft dysfunction; PRBCs, packed red blood cells; R, recipient. *Fisher exact test.
In conclusion, our study showed that a longer TBDACC was associated with lower odds of developing PGD. In terms of preventing PGD, waiting a couple of days after brain death for procurement of the heart seem to be more beneficial. Future larger prospective studies need to be performed to establish the optimal wait time after brain death for heart transplantation and evaluate the overall outcomes of patients based on their TBDACC. ACKNOWLEDGMENTS The authors thank Dr Mohammad H. Rahbar for his help in the statistical analysis of this manuscript.
Table 3. Multivariate Analysis Odds Ratio
Downtime* (<10 min) Opiates* Donor sex* (male) TBDACC 3 d* TBDACC 5 d* Inotrope* score Mean* vasopressin dose Mean* phenylephrine dose Anoxia*
0.102 4.796 2.960 0.098 0.092 0.988 0.845 1.056 4.051
95% CI
(0.007, (0.675, (0.549, (0.021, (0.010, (0.954, (0.229, (0.369, (0.846,
1.437) 34.0) 16.0) 0.443) 0.411) 1.023) 3.116) 3.023) 19.4)
P
.091 .117 .207 .0026 .0017 .513 .800 .918 .080
Abbreviations: CI, confidence interval; TBDACC, time from brain death to aortic cross clamp. *Logistic regression.
REFERENCES [1] Hunt SA. Taking heartecardiac transplantation past, present, and future. N Engl J Med 2006;355:231. [2] Goldsmith KA, Demiris N, Gooi JH, et al. Life-years gained by reducing donor heart ischemic times. Transplantation 2009;87:243. [3] Khush KK, Valantine HA. New developments in immunosuppressive therapy for heart transplantation. Expert Opin Emerg Drugs 2009;14:1. [4] Hunt SA, Haddad F. The changing face of heart transplantation. J Am Coll Cardiol 2008;52:587e98. [5] Caves PK, Stinson EB, Graham AF, et al. Percutaneous transvenous endomyocardial biopsy. JAMA 1973;225:288. [6] Kobashigawa J, Zuckermann A, Macdonald P, et al. Report from a consensus conference on primary graft dysfunction after cardiac transplantation. J Heart Lung Transplant 2014;33:327. [7] Gaies MG, Gurney JG, Yen AH, et al. Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass. Pediatr Crit Care Med 2010;11:234. [8] Wernovsky G, Wypij D, Jonas RA, et al. Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants. A comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation 1995;92: 2226e35. [9] D’Ancona G, Santise G, Falletta C, et al. Primary graft failure after heart transplantation: the importance of donor pharmacological management. Transplant Proc 2010;42:710e2. [10] D’Alessandro C, Golmard JL, Barreda E, et al. Predictive risk factors for primary graft failure requiring temporary extracorporeal membrane oxygenation support after cardiac transplantation in adults. Eur J Cardiothorac Surg 2011;40:962e9. [11] Russo MJ, Iribarne A, Hong KN, et al. Factors associated with primary graft failure after heart transplantation. Transplantation 2010;90:444e50. [12] Pratschke J, Wilhelm MJ, Kusaka M, et al. Brain death and its influence on donor organ quality and outcome after transplantation. Transplantation 1999;67:343e8. [13] Wilhelm MJ, Pratschke J, Laskowski IA, et al. Brain death and its impact on the donor heart-lessons from animal models. J Heart Lung Transplant 2000;19:414e8. [14] Shivalkar B, Van Loon J, Wieland W, et al. Variable effects of explosive or gradual increase of intracranial pressure on myocardial structure and function. Circulation 1993;87: 230e9. [15] Ryan JB, Hicks M, Cropper JR, et al. Functional evidence of reversible ischemic injury immediately after the sympathetic storm associated with experimental brain death. J Heart Lung Transplant 2003;22:922e8. [16] D’Amico TA, Meyers CH, Koutlas TC, et al. Desensitization of myocardial beta-adrenergic receptors and deterioration of left ventricular function after brain death. J Thorac Cardiovasc Surg 1995;110:746e51. [17] Atkinson C, Floerchinger B, Qiao F, et al. Donor brain death exacerbates complement- dependent ischemia/reperfusion injury in transplanted hearts. Circulation 2013;127:1290e9. [18] Cantin B, Kwok BW, Chan MC, et al. The impact of brain death on survival after heart transplantation: time is of the essence. Transplantation 2003;76:1275e9. [19] Nijboer WN, Moers C, Leuvenink HG, et al. How important is the duration of the brain death period for the outcome in kidney transplantation? Transpl Int 2011;24:14e20. [20] Bos EM, Schuurs TA, Kraan M, et al. Renal expression of heat shock proteins after brain death induction in rats. Transplant Proc 2005;37:359. [21] Avlonitis VS, Wigfield CH, Golledge HD, et al. Early hemodynamic injury during donor brain death determines the severity of primary graft dysfunction after lung transplantation. Am J Transplant 2007;7:83.
Center for Advanced Heart Failure, University of Texas Health Science Center at Houston/Memorial Hermann Hospital, Texas Medical Center, bDivision of Clinical and Translational Sciences, Department of Internal Medicine, cDivision of Critical Care, Pulmonary and Sleep, University of Texas Medical School at Houston, and dDepartment of Pharmacy, Memorial Hermann e Texas Medical Center, Houston, Texas
ABSTRACT Background. Primary graft dysfunction (PGD) is a frequent complication after cardiac transplantation and remains one of the leading causes of mortality in these patients. The objective of this case-control study is to identify donor and surgical procedure’s factors associated with PGD, and further guide possible strategies to prevent PGD. Methods. Retrospective analysis of the medical records of patients who underwent cardiac transplantation at Memorial Hermann Hospital at Texas Medical Center between October 2012 and February 2015. Results. The study population included 99 patients, of which 18 developed PGD. Univariate analysis of donor characteristics revealed opioid use (P ¼ .049) and death owing to anoxia (P ¼ .021) were associated with PGD. The recipient/donor blood type match AB/A was significantly associated with PGD (P ¼ .031). Time from brain death to aortic cross clamp (TBDACC) of 3 and 5 days were also found to be associated with PGD (P ¼ .0011 and .0003, respectively). Multivariate analysis confirmed that patients with a time from brain death to aortic cross clamp 3 and 5 days had lesser odds of developing PGD (odds ratio, 0.098 [P ¼ .0026] and OR, 0.092 [P ¼ .0017], respectively]. Conclusions. Our study showed that a longer time from brain death to aortic cross clamp was associated with lower odds of developing PGD. Therefore, postponing heart procurement for a few days after brain death seems to be beneficial in preventing PGD.
S
INCE the first cardiac transplantation in 1967, patient outcomes have improved significantly as a consequence of new surgical techniques, more effective immunosuppressive agents, better organ preservation, prophylaxis against opportunistic infections, and early rejection recognition with endomyocardial biopsies [1e5]. Nonetheless, primary graft dysfunction (PGD) remains a leading cause of 30-day mortality in these patients. The true incidence of PGD is hard to estimate owing to reporting variability by different heart transplant centers and the previous lack of a standardized definition. In April 2013, a consensus conference on PGD developed guidelines to define, diagnose, and treat patients with PGD after cardiac transplantation. This consensus included data from 47 transplant centers, comprising a
total of 9901 patients who underwent heart transplants. Among those patients, 733 (7.4%) developed PGD. Of those, 30-day and 1-year mortality was 30% and 35%, respectively. The leading reported causes of death were multiorgan system failure (70%), primary graft failure (20%), and sepsis (10%). PGD was defined as left ventricle (LV), biventricular (BV), or right ventricle (RV) failure that occurs within 24 hours after cardiac transplantation, not attributable to a discernible etiology such
*Address correspondence to Raymundo Alain QuintanaQuezada, MD, University of Texas Health Science Center at Houston, 6410 Fannin St, Suite 920, Houston, Texas 77030. E-mail: [email protected]
0041-1345/16 http://dx.doi.org/10.1016/j.transproceed.2016.02.073
ª 2016 Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
2168
Transplantation Proceedings, 48, 2168e2171 (2016)
PRIMARY GRAFT DYSFUNCTION
as hyperacute rejection, pulmonary hypertension, or a surgical complication, events that would be considered secondary graft dysfunction [6]. The increasing number of patients active on the waiting list for cardiac transplantation, the potential complications that ventricular support devices may be associated with, and the currently limited donor availability may cause the donor selection process to become more lenient. The objective of this case-control study is to identify donor and surgical procedural factors associated with PGD and to further guide possible strategies to prevent PGD.
2169
Statistical Techniques Continuous variables were reported as mean values standard deviation or median values (interquartile range) as appropriate, and compared using the Student t test or the ManneWhitney U test. To compare categorical variables, the c2 test was used or Fisher exact test if expected frequencies were <5. All variables determined to be clinically important or suggested by univariate analysis to be significant (P < .05) were included in the multivariate regression. Multivariable logistic regression analysis was performed to assess the simultaneous effect of multiple variables on PGD. The 95% confidence interval (CI) and odds ratio (OR) were reported for each variable. All statistical analyses were performed using SAS software (version 9.3, SAS Institute, Cary NC).
MATERIAL AND METHODS We retrospectively analyzed the medical records of all patients who underwent cardiac transplantation at Memorial Hermann Hospital in the Texas Medical Center between October 2012 and February 2015. PGD was defined as LV, BV, or RV failure that occurred within 24 hours after cardiac transplantation and was not attributable to a surgical complication; this was done according to the definition established by the consensus conference on PGD after cardiac transplantation [6]. Patients with secondary graft dysfunction were excluded; hyperacute rejection was ruled out by performing endomyocardial biopsies, and we did not include patients who developed RV failure owing to elevated pulmonary vascular resistance. Patients with PGD affecting the LV (PGD-LV) were further classified into 3 grades of PGD based on the echocardiographic data, hemodynamic data, and the type of support (pharmacologic or mechanical) required. Mild PGD-LV included patients with an LV ejection fraction of 40% or right atrial pressure of >15 mm Hg, pulmonary capillary wedge pressure of >20 mm Hg, and a cardiac index of <2.0 L/min/m2 (for >1 hour) who required low-dose inotropic support (inotrope score < 10). Moderate PGD-LV included patients with an LV ejection fraction of 40% or right atrial pressure of >15 mm Hg, pulmonary capillary wedge pressure of >20 mm Hg, and cardiac index of <2.0 L/min/m2 (for >1 hour) requiring high-dose inotropic support (inotrope score > 10); and placement of an intraaortic balloon pump. Severe PGD-LV included patients dependent on LV or BV mechanical circulatory support including extracorporeal membrane oxygenation, an LV assist device, BV assist device, or percutaneous LV assist device not including intraaortic balloon pump. PGD-RV is often harder to quantify; thus, no severity grading system was established. The diagnosis of PGD-RV was established in patients with a right atrial pressure of >15 mm Hg, pulmonary capillary wedge pressure of <15 mm Hg, cardiac index of <2.0 L/min/m2, and transpulmonary gradient of <15 mm Hg and/or pulmonary artery systolic pressure of <50 mm Hg or in any patient requiring RV assist device placement [6]. Donor and recipient characteristics were studied in patients who developed PGD and then compared with patients without PGD (control group). Drug dosages were converted to micrograms per kilogram per minute and the inotrope score was calculated according to the following formula: dopamine (dose 1) þ dobutamine (dose 1) þ epinephrine (dose 100) þ norepinephrine (dose 100) þ isoproterenol (dose 100) þ milrinone (dose 15) [7,8]. The doses of vasopressors, phenylephrine and vasopressin were converted to micrograms per kilogram per minute and units per kilogram per minute, respectively, and analyzed separately. Data collection, analysis, and reporting were approved by the institutional review board.
RESULTS
We reviewed a total of 102 charts. Three patients were not included in our study owing to incomplete donor records. We found 18 patients who developed PGD, and 81 served as controls. Among the patients with PGD, 4 developed PGDRV, 9 had severe PGD-LV, 2 had moderate PGD-LV, and 3 had mild PGD-LV. The donor and surgical procedure characteristics are shown in Tables 1 and 2, respectively. Among the donor characteristics, a history of opioid use (P ¼ .049) and death owing to anoxia (P ¼ .021) were associated significantly with PGD. The recipient/donor blood type match AB/A was significantly associated with PGD (P ¼ .031). Time from brain death to aortic cross clamp (TBDACC) of 3 and 5 days were also associated with the development of PGD (P ¼ .0011 and .0003, respectively). Multivariate analysis (Table 3) confirmed that patients with a TBDACC 3 days (OR, 0.098; 95% CI, 0.021e0.443; P ¼ .0026) and 5 days (OR, 0.092; 95% CI, 0.010e0.411; P ¼ .0017) had progressively lower odds of developing PGD. DISCUSSION
In our study, the rate of PGD was 18.18%, which is higher than the reported incidence in the consensus conference on PGD [6]. Upon reviewing the donor characteristics, we were unable to find a difference in the number of blood transfusions, inotrope score, phenylephrine dose, vasopressin dose, preharvest donor down time, and ischemic time between the PGD and control group. These variables are wellknown to be associated with PGD [9e11]. However, we believe our sample size was not large enough to detect a significant difference. Regarding causes of death, despite previous reports describing trauma as an independent risk factor for the occurrence of primary graft failure [10], our study showed an association between anoxia and PGD (P > .021), which disappeared in the multivariate analysis (OR, 4.051; 95% CI, 0.846e19.4; P ¼ .080). Multiple animal and human studies have shown that brain death induces release of catecholamines, cytokines and biologic mediators which, through a series of molecular events, cause deterioration of myocardial contractility and predispose the heart to ischemic injury [12e17]. Cantin et al. [18]
2170
QUINTANA-QUEZADA, RAJAPREYAR, POSTALIAN-YRAUSQUIN ET AL Table 1. Univariate Analysis of Donor’s Clinical Factors Donor Factors
Without PGD, n/N (%)
Gender difference (female)* Age (y) <20* 20e29* 30e39* 40e49* BMI (kg/m2) <20* 20e24* 25e29* 30e34* >35* Diabetes mellitus* Hypertension* CPR* Downtime < 10 min* Downtime >10 min* TBDACC < 5 d* TBDACC > 5 d* TBDACC < 3 d* TBDACC > 3 d* Marijuana use* Benzodiazepine use* Cocaine use* Amphetamine use* Opioid use* Phencyclidine use* Nonsmokers* Smokers (1e10 packs-year)* Smokers (>10 packs-year)* Mean inotropic score† Phenylephrine dose (mg/kg/min), mean SD† Vasopressin dose, mean SD (U/kg/min)† Troponin I < 1* Troponin I > 1* Ejection fraction (>50%)* Prolonged QTc Anoxia* Traumatic brain injury* Intracerebral* hemorrhage Gunshot wound*
21/81 (25.92%) 12/81 35/81 25/81 9/81
(14.81) (43.21) (30.86) (11.11)
13/81 (16.05) 23/81 (28.40) 28/81 (34.57) 12/81 (14.81) 5/81 (6.17) 3/81 (3.70) 10/81 (12.35) 25/81 (30.86) 67/81 (82.72) 14/81 (17.28) 11/81 (13.58) 70/81 (86.42) 28/81 (34.57) 54/81 (66.67) 27/81 (33.33) 11/81 (13.58) 4/81 (4.94) 8/81 (9.88) 8/81 (9.88) 1/81 (1.23) 48/81 (59.26) 25/81 (30.86) 8/81 (9.88) 16.8 0.341 0.583 0.0004 (0.003) 60/81 (74.07) 22/81 (27.16) 68/81 (83.95) 31/81 (38.27) 13/81 (16.05) 23/81 (28.4) 16/81 (19.75) 16/81 (19.75)
PGD, n/N (%)
2/18 (11.11) 2/18 10/18 3/18 3/18
P
.231
(11.11) (55.56) (16.67) (16.67)
.99 .311 .384 .444
1/18 (5.56) 6/18 (33.33) 7/18 (38.89) 3/18 (16.67) 1/18 (5.56) 2/18 (11.11) 3/18 (16.67) 8/18 (44.44) 11/18 (61.11) 7/18 (38.89) 10/18 (55.55) 8/18 (44.44) 14/18 (77.78) 3/18 (16.67) 2/18 (11.11) 3/18 (16.67) 2/18 (11.11) 4/18 (22.22) 5/18 (27.78) 0/18 (0) 11/18 (61.11) 4/18 (22.22) 3/18 (16.67) 11.7 0.335 (0.681) 0.001 (0.004) 10/18 (55.56) 7/18 (38.89) 17/18 (94.44) 9/18 (50) 8/18 (44.44) 2/18 (11.11) 6/18 (33.33) 2/18 (11.11)
.454 .775 .787 .731 .99 .219 .699 .277 .55 .0003 .0011 .574 .156 1 .221 .049 1 1 .577 .411 .254 .990 .338 .384 .294 .285 .021 .227 .217 .515
Abbreviations: BMI, body mass index; CPR, cardiopulmonary resuscitation; TBDACC, time from brain death to aortic cross clamp; QTc, corrected QT interval; SD, standard deviation. *Fisher exact test. † 2 c test.
studied 475 heart transplant cases and found that patients with a shorter management time; defined as time from an incident leading to brain death (ie, trauma, subarachnoid hemorrhage) until aortic cross clamp, had an increased survival when compared with patients with a shorter management time. When they specifically evaluated the TBDACC, they did not find a significant association with survival. The authors concluded that events before the time point when the patient developed brain death somehow damaged the donor heart [18]. In our study, we found that patients with a TBDACC of 3 days had significantly lower odds of developing PGD compared to patients with a TBDACC of <3 days.
When we analyzed the patients with a TBDACC of 5 days we found that their odds of developing PGD were even lower. Similar findings have been described in renal transplant cases, in which a longer TBDACC decreased the risk of delayed graft function [19]. The proposed explanation, based on a few animal studies done in kidney [20] and lung [21] transplant models, is that if more time elapses after brain death, adaptive tissue mechanisms to recover from the initial catecholamine surge injury can take place. The study is limited because it is retrospective and we are not considering other complications of heart transplantation, including hyperacute rejection episodes and surgical complications.
PRIMARY GRAFT DYSFUNCTION
2171
Table 2. Univariate Analysis of Surgical Procedure’s Factors Surgical procedure factors
Without PGD
Gender matches* (recipient-donor) Male to male 42/81 (51.85%) Male to female* 19/81 (23.46%) Female to female* 9/81 (11.11%) Female to male* 13/81 (16.05%) CMV status* R ()/D(þ) 18/81 (22.22%) R()/D() 16/81 (19.75%) R(þ)/D() 12/81 (14.81%) R(þ)/D(þ) 35/81 (43.21%) Ischemic time (h)* <2 13/81 (16.05%) 2e4 59/81 (72.84%) >4 9/81 (11.11%) Blood type matches (R/D)* O/O 31/81 (38.27%) A/O 4/81 (4.94%) B/O 5/81 (6.17%) A/A 28/81 (34.57%) B/B 10/81 (12.35%) AB/AB 2/81 (2.47%) AB/A 0/81 (0%) AB/B 1/81 (1.23%) PRBCs transfusion >10* 12/81 (14.81%)
PGD
P
14/18 2/18 0/18 2/18
(77.78%) (11.11%) (0%) (11.11%)
.065 .348 .357 .999
1/18 4/18 4/18 9/18
(5.56%) (22.22%) (22.22%) (50%)
.0477 .751 .183 .613
2/18 (11.11%) 13/18 (72.22%) 3/18 (16.67%) 6/18 0/18 1/18 6/18 3/18 0/18 2/18 0/18 3/18
(33.33%) (0%) (5.55%) (33.33%) (16.67%) (0%) (11.11%) (0%) (16.67%)
.9
.791 .999 .999 .999 .698 .999 .031 .999 .731
Abbreviations: CMV, cytomegalovirus; D, donor; PGD, primary graft dysfunction; PRBCs, packed red blood cells; R, recipient. *Fisher exact test.
In conclusion, our study showed that a longer TBDACC was associated with lower odds of developing PGD. In terms of preventing PGD, waiting a couple of days after brain death for procurement of the heart seem to be more beneficial. Future larger prospective studies need to be performed to establish the optimal wait time after brain death for heart transplantation and evaluate the overall outcomes of patients based on their TBDACC. ACKNOWLEDGMENTS The authors thank Dr Mohammad H. Rahbar for his help in the statistical analysis of this manuscript.
Table 3. Multivariate Analysis Odds Ratio
Downtime* (<10 min) Opiates* Donor sex* (male) TBDACC 3 d* TBDACC 5 d* Inotrope* score Mean* vasopressin dose Mean* phenylephrine dose Anoxia*
0.102 4.796 2.960 0.098 0.092 0.988 0.845 1.056 4.051
95% CI
(0.007, (0.675, (0.549, (0.021, (0.010, (0.954, (0.229, (0.369, (0.846,
1.437) 34.0) 16.0) 0.443) 0.411) 1.023) 3.116) 3.023) 19.4)
P
.091 .117 .207 .0026 .0017 .513 .800 .918 .080
Abbreviations: CI, confidence interval; TBDACC, time from brain death to aortic cross clamp. *Logistic regression.
REFERENCES [1] Hunt SA. Taking heartecardiac transplantation past, present, and future. N Engl J Med 2006;355:231. [2] Goldsmith KA, Demiris N, Gooi JH, et al. Life-years gained by reducing donor heart ischemic times. Transplantation 2009;87:243. [3] Khush KK, Valantine HA. New developments in immunosuppressive therapy for heart transplantation. Expert Opin Emerg Drugs 2009;14:1. [4] Hunt SA, Haddad F. The changing face of heart transplantation. J Am Coll Cardiol 2008;52:587e98. [5] Caves PK, Stinson EB, Graham AF, et al. Percutaneous transvenous endomyocardial biopsy. JAMA 1973;225:288. [6] Kobashigawa J, Zuckermann A, Macdonald P, et al. Report from a consensus conference on primary graft dysfunction after cardiac transplantation. J Heart Lung Transplant 2014;33:327. [7] Gaies MG, Gurney JG, Yen AH, et al. Vasoactive-inotropic score as a predictor of morbidity and mortality in infants after cardiopulmonary bypass. Pediatr Crit Care Med 2010;11:234. [8] Wernovsky G, Wypij D, Jonas RA, et al. Postoperative course and hemodynamic profile after the arterial switch operation in neonates and infants. A comparison of low-flow cardiopulmonary bypass and circulatory arrest. Circulation 1995;92: 2226e35. [9] D’Ancona G, Santise G, Falletta C, et al. Primary graft failure after heart transplantation: the importance of donor pharmacological management. Transplant Proc 2010;42:710e2. [10] D’Alessandro C, Golmard JL, Barreda E, et al. Predictive risk factors for primary graft failure requiring temporary extracorporeal membrane oxygenation support after cardiac transplantation in adults. Eur J Cardiothorac Surg 2011;40:962e9. [11] Russo MJ, Iribarne A, Hong KN, et al. Factors associated with primary graft failure after heart transplantation. Transplantation 2010;90:444e50. [12] Pratschke J, Wilhelm MJ, Kusaka M, et al. Brain death and its influence on donor organ quality and outcome after transplantation. Transplantation 1999;67:343e8. [13] Wilhelm MJ, Pratschke J, Laskowski IA, et al. Brain death and its impact on the donor heart-lessons from animal models. J Heart Lung Transplant 2000;19:414e8. [14] Shivalkar B, Van Loon J, Wieland W, et al. Variable effects of explosive or gradual increase of intracranial pressure on myocardial structure and function. Circulation 1993;87: 230e9. [15] Ryan JB, Hicks M, Cropper JR, et al. Functional evidence of reversible ischemic injury immediately after the sympathetic storm associated with experimental brain death. J Heart Lung Transplant 2003;22:922e8. [16] D’Amico TA, Meyers CH, Koutlas TC, et al. Desensitization of myocardial beta-adrenergic receptors and deterioration of left ventricular function after brain death. J Thorac Cardiovasc Surg 1995;110:746e51. [17] Atkinson C, Floerchinger B, Qiao F, et al. Donor brain death exacerbates complement- dependent ischemia/reperfusion injury in transplanted hearts. Circulation 2013;127:1290e9. [18] Cantin B, Kwok BW, Chan MC, et al. The impact of brain death on survival after heart transplantation: time is of the essence. Transplantation 2003;76:1275e9. [19] Nijboer WN, Moers C, Leuvenink HG, et al. How important is the duration of the brain death period for the outcome in kidney transplantation? Transpl Int 2011;24:14e20. [20] Bos EM, Schuurs TA, Kraan M, et al. Renal expression of heat shock proteins after brain death induction in rats. Transplant Proc 2005;37:359. [21] Avlonitis VS, Wigfield CH, Golledge HD, et al. Early hemodynamic injury during donor brain death determines the severity of primary graft dysfunction after lung transplantation. Am J Transplant 2007;7:83.