Author's Accepted Manuscript

Elevated Pre-Transplant Pulmonary Vascular Resistance is Not Associated with Mortality in Children without Congenital Heart Disease: A Multi-Center Study Marc E. Richmond MD, MS, Yuk M. Law MD, Bibhuti B. Das MD, Melanie D. Everitt MD, Manisha Kukreja MD, MPH, David C. Naftel PhD, Mariska S. Kemna MD, Heather T. Henderson MD, Kimberly Beddows MS, CPNP, F. Jay Fricker MD, William T. Mahle MD, on behalf of the Pediatric Heart Transplant Study Investigators

PII: DOI: Reference:

S1053-2498(14)01119-X http://dx.doi.org/10.1016/j.healun.2014.04.021 HEALUN5766

To appear in:

J Heart Lung Transplant

http://www.jhltonline.org

Cite this article as: Marc E. Richmond MD, MS, Yuk M. Law MD, Bibhuti B. Das MD, Melanie D. Everitt MD, Manisha Kukreja MD, MPH, David C. Naftel PhD, Mariska S. Kemna MD, Heather T. Henderson MD, Kimberly Beddows MS, CPNP, F. Jay Fricker MD, William T. Mahle MD, on behalf of the Pediatric Heart Transplant Study Investigators, Elevated Pre-Transplant Pulmonary Vascular Resistance is Not Associated with Mortality in Children without Congenital Heart Disease: A Multi-Center Study, J Heart Lung Transplant, http://dx.doi.org/10.1016/j.healun.2014.04.021 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Elevated Pre-Transplant Pulmonary Vascular Resistance is Not Associated with Mortality in Children without Congenital Heart Disease: A Multi-Center Study Marc E Richmond, MD, MS1, Yuk M Law, MD6, Bibhuti B. Das, MD2, Melanie D Everitt, MD3, Manisha  Kukreja, MD, MPH5, David C Naftel, PhD5, Mariska S. Kemna, MD6, Heather T. Henderson, MD7, Kimberly  Beddows, MS, CPNP1, F Jay Fricker, MD4 and William T. Mahle, MD8 on behalf of the Pediatric Heart  Transplant Study Investigators    1

Division of Pediatric Cardiology, Columbia University College of Physicians and Surgeons, New York, NY, 

United States, 10032; 2Division of Cardiology, Children's Medical Center, UT Southwestern Medical  Center, Dallas, TX, United States, 75235; 3Division of Cardiology, Primary Children's Hospital, Salt Lake  City, Utah, United States, 84113; 4Division of Pediatric Cardiology, University of Florida, Gainsville, FL,  United States, 32610; 5Cardiothoracic Surgery, University of Alabama at Birmingham, Birmingham, AL,  United States, 35294; 6Seattle Children's Hospital, Seattle, WA, United States; 7Division of Pediatric  Cardiology, Duke University School of Medicine, Durham, NC, United States and 88Children’s Healthcare  of Atlanta and Emory University School of Medicine, Atlanta, GA, United States.

ABSTRACT Traditionally, an elevated pulmonary vascular resistance index (PVRI) has been a relative  contraindication to pediatric heart transplantation (HT). This study examined the risk of elevated  pretransplant PVRI upon early (30‐day) and intermediate‐term mortality in pediatric HT recipients  without congenital heart disease (CHD). 

1   

A review of the prospective multi‐center Pediatric Heart Transplant Study registry identified all patients  without CHD in whom a pre‐HT PVRI was recorded. Of 35 participating centers, 29 reported OHT in  children with a markedly elevated PVRI (>5 WU x m2, corresponding to the highest quartile). Multiphase  parametric analysis was performed adjusting for potential risk factors to assess PVRI’s association with  early and intermediate term mortality.  Between 1993 and 2011, 1909 children without CHD underwent OHT at a median age of 9.7 years (range  1.6 months ‐18 years). Of those, 795 (42%) had a recorded or calculable pre‐HT PVRI, and PVRI >5 WU x  m2 was present in 193 (24%) patients. For all recipients, median pre‐HT PVRI was 3.15 WU x m2 (0.4 to  23); 2.8 in infants 10 years (p=0.03).  Multivariable hazard analysis controlling for graft ischemic time and pre‐HT ventilation showed no  association of elevated PVRI with early mortality (RR=1.2, p=0.66), nor with intermediate mortality when  controlled for year of HT, age, race, and presensitization (RR=0.7, p=0.27).   In this large, multi‐center cohort of pediatric HT recipients without CHD, elevation of PVRI did not affect  post‐HT survival, suggesting that the barrier of elevated PVRI can be successfully overcome in this  population. 

Introduction Heart transplantation continues to be the most successful long-term therapy for children with end-stage heart failure from cardiomyopathy. Survival after transplantation has been linked to multiple pretransplant risk factors including pulmonary vascular resistance index (PVRI) (1, 2). A PVRI greater than 5 or 6 WU x m2 has historically been considered a relative contraindication to heart transplantation in pediatric patients due to an increased risk of postoperative right ventricular dysfunction and early mortality (3, 4). However, more recent single center studies suggest that current practice may be too 2   

restrictive with regards to PVRI, and that excellent outcomes can occur in patients with PVRI above these traditional thresholds (3-5). Despite increasing treatment options in the current era for lowering pulmonary vascular resistance, there remains controversy over what level of PVRI, if any, should be considered as a contraindication to orthotopic heart transplant. Clinical practice variation across individual centers likely contributes to the disparity of outcomes reported to date in single center studies of transplant outcomes in children with pulmonary hypertension. As such, the aim of the current analysis is to determine the impact of pretransplant PVRI upon post-transplant survival in a large, multicenter registry of pediatric heart transplant recipients, and to assess for trends over time and across centers with regards to candidate selection for heart transplantation in children with pulmonary hypertension. Methods Study Population/Design Data from the Pediatric Heart Transplant Study (PHTS) database, a prospective registry of pediatric (≤18 years) heart transplant recipients from 35 institutions across North America and the United Kingdom were utilized for this analysis. Details of data collection have been previously published (6) and a list of participating centers is provided in the Appendix. A query of this database was performed to identify all children who underwent isolated heart transplantation for an indication other than congenital heart disease between January 1, 1993 and December 31, 2011. Demographics, pretransplant hemodynamics, pretransplant therapeutics and intraoperative factors were collected. PVRI was collected or calculated from available hemodynamic data and reported in indexed Woods Units (WU x m2). PVRI was calculated utilizing mean pulmonary artery pressure (mPAP), Pulmonary Capillary Wedge Pressure (PCWP) and Cardiac Index (CI)) entered into the equation: ‫ ܫ‬ൌ

௠௉஺௉ି௉஼ௐ௉ ஼ூ

. All hemodynamic data

collected in the PHTS database, including PVRI are reported as “best hemodynamics” at the time of cardiac catheterization, and are collected on forms at both time of listing for heart transplant and at the 3   

time of transplantation. For all identified patients with more than one pretransplant PVRI reported, the value closest to the time of transplant was used. If PVRI was not recorded, unable to be calculated, or was obtained greater than one year prior to transplantation, the patient was excluded from this analysis. The primary outcome variable was defined as all-cause mortality and was analyzed as both a continuous variable as well as dichotomized at 30-days. Statistical Analysis All continuous variables were examined for normality and are expressed as mean ± standard deviation, or if nonnormal, as the median with either the range or IQR. Initial evaluation with ROC curve and Cox proportional hazard modeling was performed to identify a cut-off value for elevated PVRI and factors were compared between the PVRI groups using independent sample Students T-test, Mann Whitney U or Chi-squared analysis as appropriate. Univariate logistic regression was used to evaluate the association between PVRI and 30-day mortality. Parametric 3-phase hazard modeling was performed to identify variables associated with early, intermediate, and late phase risk of mortality (7). PVRI as a predictor variable was forced into the final model regardless of statistical significance. All statistical tests were two-sided and a p-value of 0.05 was considered significant. Statistical analyses were performed using SAS 9.3 software for UNIX and were performed at the PHTS Data Coordinating Center at University of Alabama-Birmingham.

Results Study Cohort A total of 3769 children in the PHTS database underwent heart transplantation between 1993 and 2011. Of these, 1851 carried a diagnosis of congenital heart disease and 9 were multi-organ transplant recipients and were subsequently excluded from the study. Among the 1909 children who met study inclusion

4   

criteria, a recorded PVRI within one year prior to transplant was available for 639 patients and was able to be calculated for an additional 156 patients resulting in a total of 795 patients available for analysis. This study cohort of 795, represented 42% of the potential subject population as 1114 (58%) were excluded due to lack of available PVRI data. The demographics of the 795 children who comprised the analytic cohort are as follows: 48% were male; the median age at transplant was 9.7 years (0.13-18 years); and the indications for transplant were idiopathic dilated cardiomyopathy in 56%, restrictive cardiomyopathy in 15%, hypertrophic cardiomyopathy in 4%,other, including myocarditis, in 11%, and not specified (but not CHD) in 14%. Mean waitlist time was 75.5 days with 83.8% listed as Status 1A at the time of transplant, and 13% on mechanical circulatory support at the time of transplant (77 on ventricular assist devices, 24 on ECMO). Population Characteristics For the overall cohort, the median pretransplant PVRI was 3.15 WU X m2 (Range: 0.4 to 23). Additionally, these subjects had a median TPG of 8.0 mmHg (IQR 6.0-12.0) and a median cardiac index of 2.7 L/min/ (2.1-3.5). At the time of hemodynamic data collection 366 (46%) subjects were on at least one continuous vasoactive inhaled agent or intravenous infusion: 21% (n=169) on milrinone; 15% (n=121) on 100% Oxygen; 13% (n=104) on dobutamine; 7% (n=59) on inhaled Nitric Oxide; and 7% (n=54) on nitroprusside. As ROC analysis did not reveal a suitable inflection point for 30-day mortality, the cumulative distribution curve of PVRI was examined (Figure 1). Regression analysis at PVRI intervals of 0.2 WU x m2 was then performed. Sample analyses at 1, 5, 10, 11, and 12 WU x m2 had p-values of 0.31, 0.97, 0.74, 0.86, and 0.69 respectively, revealing no identifiable breakpoint as associated with mortality. As shown in Figure 1, a PVRI of 5 WU x m2 represented the 75th percentile and a PVRI of 10 WU x m2 corresponded to approximately the 95th percentile. Kaplan-Meier analysis among these three PVRI groups (10 WU x m2) revealed no significant differences in survival with a

5   

log-rank p-value of 0.93. Therefore, in the interest of parsimony a single PVRI cut-off value of 5 WU x m2 (corresponding to the 75th percentile) was chosen resulting in 193 patients classified with markedly elevated PVRI. Twenty nine of 35 participating PHTS centers performed at least one heart transplant during the study period in a child with a PVRI in this highest quartile. When comparing children with markedly elevated PVRI to those without, there were no differences with regards to sex, race, or weight. There was a small difference between groups with respect to age at transplant (median age 10 years vs. 8 years for those with and without markedly elevated PVRI, respectively) (Table 1a). However, when grouped by age category, younger children (1-10yrs) had a somewhat higher median PVRI (3.5 WU x m2) versus older (>10 years) children (3.0 WU x m2) and infants (2.8 WU x m2),p=0.03. When PVRI was dichotomized and age groups were compared, there was no difference in the frequency of markedly elevated PVRI (>5 WU x m2) (Table 1a). When the PVRI groups were assessed for differences in risk factors previously associated with transplant mortality, there was no difference in use of ECMO, VAD or mechanical ventilation at the time of transplant, or in total cardiopulmonary bypass time or graft ischemic time (Table 1a). There was no difference in donor age or donor weight based on high PVRI. Additionally, there was no difference in the donor to recipient weight ratio, suggesting that patients with elevated PVRI were not preferentially being transplanted with oversized donors. Patients with an elevated PVRI did have a lower median creatinine as compared to patients without elevated PVRI (0.5 vs 0.6, p5 WU x m2) Figure 3. Comparison of cumulative distribution functions of age between noncongenital heart disease patients who had PVRI data available vs. those who did not. Figure 4. Comparison of survival between noncongenital heart disease patients who had PVRI data available vs. those who did not. Figure 5. Distribution of patients with PVRI>5 WU x m2 from 1993 to 2011 as a percentage of total transplants. Figure 6. Distribution of proportion of patients transplanted with PVRI >5 WU x m2 by center volume over the study period. Tables Table 1a. Demographic Data of Study Population as compared between patients with elevated PVRI (PVRI > 5 WU x m2) and those without elevated PVRI

13   

Table 1b. Frequency of therapeutic agents used to obtain best hemodynamics on pre-transplant catheterization as compared between patients with elevated PVRI (PVRI > 5 WU x m2) and those without elevated PVRI Table 2. Multivariable hazard analysis. Of note, comparison values for continuous variables are chosen from 75th and 25th percentile values References 1.  Fukushima N, Gundry SR, Razzouk AJ, Bailey LL: Risk factors for graft failure associated with  pulmonary hypertension after pediatric heart transplantation. J Thorac Cardiovasc Surg 1994;107:985‐9.  2.  Kirk R, Dipchand AI, Edwards LB, et al.: The Registry of the International Society for Heart and  Lung Transplantation: Fifteenth Pediatric Heart Transplantation Report—2012. The Journal of Heart and  Lung Transplantation 2012;31:1065‐72.  3.  Addonizio LJMD, Gersony WMMD, Robbins RCMD, et al.: Elevated pulmonary vascular  resistance and cardiac transplantation. Circulation 1987;76 Supplement:V‐52‐V‐5.  4.  Bograd AJ, Mital S, Schwarzenberger JC, et al.: Twenty‐year experience with heart  transplantation for infants and children with restrictive cardiomyopathy: 1986‐2006. American Journal  of Transplantation 2008;8:201‐7.  5.  Chiu P, Russo MJ, Davies RR, Addonizio LJ, Richmond ME, Chen JM: What is high risk? Redefining  elevated pulmonary vascular resistance index in pediatric heart transplantation. The Journal of Heart  and Lung Transplantation 2012;31:61‐6.  6.  Dipchand AI, Kirk R, Mahle WT, et al.: Ten yr of pediatric heart transplantation: A report from  the Pediatric Heart Transplant Study. Pediatric Transplantation 2013;17:99‐111.  7.  Blackstone EH, Naftel DC, Turner ME: The Decomposition of Time‐Varying Hazard into Phases,  Each Incorporating a Separate Stream of Concomitant Information. Journal of the American Statistical  Association 1986;81:615‐24.  8.  Buddhe S, Du W, L’Ecuyer T: Impact of pulmonary hypertension on transplant outcomes in  pediatric cardiomyopathy patients. Pediatric Transplantation 2012;16:367‐72.  9.  Canter CE, Shaddy RE, Bernstein D, et al.: Indications for Heart Transplantation in Pediatric Heart  Disease. Circulation 2007;115:658‐76.  10.  Goland S, Czer LS, Kass RM, et al.: Pre‐existing pulmonary hypertension in patients with end‐ stage heart failure: impact on clinical outcome and hemodynamic follow‐up after orthotopic heart  transplantation. The Journal of Heart and Lung Transplantation 2007;26:312‐8.  11.  Klotz S, Wenzelburger F, Stypmann J, et al.: Reversible pulmonary hypertension in heart  transplant candidates: to transplant or not to transplant. The Annals of Thoracic Surgery 2006;82:1770‐ 3.  12.  Delgado JF, Gómez‐Sánchez MA, Sáenz de la Calzada C, et al.: Impact of mild pulmonary  hypertension on mortality and pulmonary artery pressure profile after heart transplantation. The  Journal of Heart and Lung Transplantation 2001;20:942‐8.  13.  Hoskote A, Carter C, Rees P, Elliott M, Burch M, Brown K: Acute right ventricular failure after  pediatric cardiac transplant: predictors and long‐term outcome in current era of transplantation  medicine. The Journal of Thoracic and Cardiovascular Surgery 2010;139:146‐53.  14.  Ardehali A, Hughes K, Sadeghi A, et al.: Inhaled nitric oxide for pulmonary hypertension after  heart transplantation. Transplantation 2001;72:638‐41. 

14   

15.  Davies RR, Russo MJ, Mital S, et al.: Predicting survival among high‐risk pediatric cardiac  transplant recipients: an analysis of the United Network for Organ Sharing database. J Thorac Cardiovasc  Surg 2008;135:147‐55.  16.  Ofori‐Amanfo G, Hsu D, Lamour JM, et al.: Heart transplantation in children with markedly  elevated pulmonary vascular resistance: Impact of right ventricular failure on outcome. The Journal of  Heart and Lung Transplantation 2011;30:659‐66.  17.  Chen JM, Levin HR, Michler RE, Prusmack CJ, Rose EA, Aaronson KD: Reevaluating the  significance of pulmonary hypertension before cardiac transplantation: Determination of optimal  thresholds and quantification of the effect of reversibility on perioperative mortality. The Journal of  Thoracic and Cardiovascular Surgery 1997;114:627‐34.  18.  Liden H, Haraldsson Å, Ricksten S‐E, Kjellman U, Wiklund L: Does pretransplant left ventricular  assist device therapy improve results after heart transplantation in patients with elevated pulmonary  vascular resistance? Eur J Cardiothorac Surg 2009;35:1029‐35.  19.  Chang PP, Longenecker JC, Wang N‐Y, et al.: Mild vs severe pulmonary hypertension before  heart transplantation: different effects on posttransplantation pulmonary hypertension and mortality.  The Journal of Heart and Lung Transplantation 2005;24:998‐1007. 

15   

Fig 1

16   

Fig 2

17   

Fig 3

18   

Fig 4

19   

Fig 5

20   

Fig 6

21   

Table 1a PVRI ≤5 WU x m2 (n=602) 10 (0.2-18)

PVRI >5 WU x m2 (n=193) 8 (0.1-18)

10 years

308 (51%)

82 (42%)

296 (49%) 135 (22%) 29 (3-130) 10 (0.02-45) 37 (3-130) 1.24 (0.43-3.8)

89 (46%) 50 (26%) 21 (3-110) 8 (0.01-52) 29 (3-113) 1.22 (0.53-2.77)

1993-1998

161 (27%)

39 (20%)

1999-2004

180 (30%)

73 (38%)

>2004

261 (43%)

81 (42%)

18 (3%) 56 (9%) 76 (13%) 0.6 (0.1-16) 198 (40-538) 116 (0-351)

6 (3%) 21 (11%) 27 (14%) 0.5 (0.1-25) 201 (51-472) 115 (0-327)

1 0.57 0.62 5 WU x m2 (n=193)

p-value

Age (years) Age Category

Male sex Race Black Weight (kg) Donor Age (years) Donor Weight (kg) D:R weight ratio (kg) Year of Transplant

ECMO Ventricular Assist Device Mechanical Ventilator Creatinine (mg/dL) Graft Ischemic Time (min) CPB Time (min)  

p-value 0.04 0.1

0.50 0.32 0.09 0.05 0.06 0.99 0.06

Table 1b 

Agents used for best hemodynamics? Yes No

0.02 263 (44%) 339 (56%)

103 (53%) 90 (47%)

100% Oxygen Dopamine Dobutamine Amiodarone Milrinone

86 (14%) 27 (4%) 77(13%) 2 (