1-Year Mortality in Adult Congenital Heart Disease Heart Transplant Recipients: Can a Risk Score Be Used To Identify Those at Highest Risk?

Abstracts S121 Conclusion: Children listed for heart transplant are commonly either underweight or obese. Underweight patients have high risk-adjusted mortality prior to transplatation.

Demographic and Clinical Variables Underweight Gender (male)

404 (57.6%)

Normal Weight Overweight

Obese

p value

1332 (57.4%)

251 (57.1%)

383 (66.8%)

0.0004

Age (y, mean±SD) 10.4±5.0

11.4±4.7

12.2±4.4

12.2±4.4

< 0.0001

Etiology: CHD

306 (43.7%)

919 (39.6%)

131 (29.8%)

146 (25.5%)

< 0.0001

Etiology: DCM

311 (44.4%)

1049 (45.2%)

249 (56.6%)

371 (64.8%)

< 0.0001

Ethnicity: African- 103 (20.5%) American

360 (21.2%)

82 (25.2%)

118 (30.0%)

< 0.0001

Ethnicity: White (non-Hispanic)

292 (58.2%)

1006 (59.3%)

172 (52.9%)

174 (44.3%)

< 0.0001

Socioeconomic Status: Low

176 (28.1%)

624 (28.7%)

150 (35.6%)

231 (41.9%)

< 0.0001

Socioeconomic Status: High

156 (24.9%)

585 (26.9%)

84 (20.1%)

78 (14.2%)

< 0.001

Status 1A/1B

458 (65.3%)

1385 (59.7%)

269 (61.1%)

362 (63.2%)

NS

Hospitalized (ICU) 258 (55.1%)

781 (54.2%)

142 (54.8%)

201 (55.5%)

NMS

Mechanical Support

78 (11.1%)

307 (13.2%)

62 (14.1%)

79 (13.8%)

NS

Albumin <  3.5

143 (39.1%)

527 (39.9%)

93 (36.6%)

153 (44.1%)

NS

Renal Insufficiency16 (3.4%)

71 (4.2%)

16 (4.9%)

27 (6.2%)

0.02

Mechanical ventilation

278 (12.0%)

54 (12.4%)

68 (11.9%)

NS

100 (14.3%)

follow-up 1.9y [IQR:35d-6y]. Predictor of 10y survival in multivariable model for infants was dilated cardiomyopathy [OR:3.0, 95% CI:1.2-7.1] while predictor of mortality within 10y was pretransplant mechanical ventilation [OR:1.7, 95%CI:1.1-2.5]. For children, male gender predicted 10y survival [OR:2.1, 95%CI:1.5-3.0] and increasing donor-recipient height ratio predicted mortality within 10y [OR:1.8, 95%CI:1.1-3.3]. For teenagers, white race [OR:2.8, 95%CI:1.7-4.5] and dilated cardiomyopathy predicted 10y survival [OR:2.1, 95%CI:1.5-3.0] (Table). Conclusion: Factors predicting long-term heart transplant survival differ for different pediatric age-groups and include diagnosis, gender, race, height and pretransplant mechanical ventilation. Understanding these age-specific factors can help clinicians make better patient selection and may help in counseling families on expectations from heart transplantation. 

3( 17)

3( 16) Predictors of Long-Term Survival after Pediatric Heart Transplantation Change with Age M.S. Khan ,1 T. Alsaied,2 F. Zafar,1 C.D. Castleberry,1 R. Bryant III,1 I. Wilmot,3 J.L. Jefferies,3 D.L. Morales.1  1Cardiothoracic Surgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; 2Pediatrics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; 3Cardiology, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH. Purpose: Over 11,000 pediatric heart transplants (HTx) have been reported worldwide with improving survival. Although the median survival continues to improve, certain patients continue to have a poor outcome. The purpose of this study was to identify factors that can predict 10-year survival after HTx for different pediatric age-groups. Methods: This was a retrospective case-control study utilizing the Organ Procurement and Transplant Network data from Oct 1987-Apr 2002 of pediatric HTx recipients (age ≤ 18y). The cases included HTx recipients who survived > 10y, and controls (patients who died within 10y of HTx). Predictors associated with 10y survival were identified separately for 3 pediatric agegroups [infants (< 1y), children (> 1-10y) and teenagers (> 10-18y)] using multivariable logistic regression model. Results: There were 1389 cases with median follow-up 14.0y [IQR:1217y] and 1577 controls (patients who died within 10y of HTx) with median

1-Year Mortality in Adult Congenital Heart Disease Heart Transplant Recipients: Can a Risk Score Be Used To Identify Those at Highest Risk? L.J. Burchill ,1 L.B. Edwards,2 J. Stehlik,3 A.I. Dipchand,4 H.J. Ross.5   1Adult Congenital Heart Disease, Oregon Health Sciences University, Portland, OR; 2International Society of Heart and Lung Transplant Registry, Addison, TX; 3University of Utah, Salt Lake City, UT; 4Hospital for Sick Children, Toronto, ON, Canada; 5Cardiac Transplant Program, Toronto General Hospital, Toronto, ON, Canada. Purpose: This study evaluated whether recipient and donor characteristics identify adult congenital heart disease (ACHD) patients at highest risk of mortality in the first year (yr) after heart transplantation (HTx). Methods: ACHD HTx mortality was evaluated using the International Society for Heart and Lung Transplantation (ISHLT) Transplant Registry. The cohort included all ACHD patients (> 18 years) undergoing HTx between January 1, 2005 and December 31, 2009. A Cox proportional hazards model was used to examine the relative risk of recipient, donor and transplant characteristics on death within 1-yr post HTx. A risk score was generated for each of the 437 ACHD transplants identified. Kaplan-Meier survival and log-rank statistic tests were used to compare survival in patients with 0, 1, 2 and ≥  3 risk factors (RFs) for 1-yr mortality. Results: Recipient age group (22-32 yrs, p< 0.01), BMI > 30g/m2 (p< 0.01), creatinine > 1.5mg/dL (p< 0.0001), female recipient/female donor pairing (p= 0.04) and donor CMV+ status (P= 0.04) were all significant predictors of mortality within 1-yr. Ischemia time >  5 hrs was borderline significant (p= 0.059). Of those included in KM survival analyses (N= 239), 20% had 0 RFs, 46% had 1 RF, and 28% had 2 RFs. ACHD HTx recipients with ≥  3 RFs were uncommon (5%). 1-yr survival decreased with increasing risk score, ranging from 96%, 82% and 70% for recipients with a risk score of 0, 1 and 2, respectively. 1-yr survival was 45% in patients with ≥  3 RFs (Figure 1).

S122

The Journal of Heart and Lung Transplantation, Vol 32, No 4S, April 2014

Conclusion: The proposed risk score appears to discriminate 1-yr survival in ACHD HTx recipients. If validated, this risk score may help in defining acceptable thresholds of risk among ACHD patients undergoing evaluation for HTx. Future research will focus on validating this risk score in an additional cohort to determine its predictive value.

3( 18) Evaluating Pediatric Heart Re-transplant Candidates: Using a Risk Prediction Model to Estimate Early Mortality after Re-Transplant S.L. Siehr ,1 S.A. Hollander,1 K. Gauvreau,2 D.N. Rosenthal,1 T.P. Singh,2 B. Kaufman,1 J. Yeh,1 C.S. Almond.2  1Pediatric Cardiology, Stanford University School of Medicine, Palo Alto, CA; 2Pediatric Cardiology, Harvard Medical School, Boston, MA. Purpose: Pediatric heart transplant (HT) centers may be reluctant to offer re-transplant to children within 6 months of primary HT because of studies suggesting graft survival is poor. The purpose of this study was to explore an alternate method for risk-stratifying re-transplant candidates by determining whether a previously validated pediatric post-transplant risk-prediction model can accurately predict early mortality in heart re-transplant candidates. Methods: All children listed for heart re-transplantation between 2004 and 2010 were identified using Organ Procurement and Transplant Network (OPTN) data. Using standard model discrimination and calibration statistics, the observed in-hospital mortality of the re-transplant cohort was compared to their predicted in-hospital mortality using a validated risk-prediction model originally developed in a cohort of primary HT candidates. Results: Of 137 children who underwent heart re-transplantation during the 6-year study period, the median age was 13 years (Interquartile range (IQR) 9, 15), the median weight was 41 kg (IQR 26, 54); 4% were supported on Extra-corporeal membrane oxygenation (ECMO) at transplant, 7% a ventilator, 34% were listed UNOS status 1A, 46% had an abnormal estimated creatinine clearance, and 13% had a bilirubin level ≥ 2 mg/dL. Overall, 9 of 137 patients (7%) died prior to hospital discharge. When the original risk prediction model was applied to the re-transplant cohort, the C-statistic (0.71) and Hosmer-Lemeshow goodness-of-fit (P= 0.37) suggested acceptable prediction of post-transplant in-hospital mortality. Conclusion: Overall unadjusted in-hospital mortality for children undergoing heart re-transplant is relatively similar to patients undergoing primary HT. A risk-prediction model validated previously in children undergoing primary HT performs reasonably well in predicting post-transplant in-hospital mortality in children undergoing heart re-transplant. This model may be a useful alternative method for assessing a child’s re-transplant candidacy independent of time since transplant. 3( 19) Prevalence of BK Polyomavirus Infection and Association with Renal Function in Pediatric Heart Transplant Recipients A.L. Ducharme-Smith ,1 A.E. Bobrowski,2 B.Z. Katz,3 C.L. Backer,4 E. Pahl.5  1Northwestern Feinberg School of Medicine, Chicago, IL; 2Nephrology, Ann & Robert H. Lurie Children’s Hospital, Chicago, IL; 3Infectious Disease, Ann & Robert H. Lurie Children’s Hospital, Chicago, IL; 4Cardiovascular-Thoracic Surgery, Ann & Robert H. Lurie Children’s Hospital, Chicago, IL; 5Medical Director, Heart Transplant Program, Ann & Robert H. Lurie Children’s Hospital, Chicago, IL. Purpose: BK polyomavirus (BKV) infection and related nephropathy is a well-known cause of renal dysfunction in renal transplant recipients; however the impact of BKV in the native kidneys of pediatric heart

transplant (HTx) recipients is unknown. Our study assessed the prevalence of BKV infection and its association with renal dysfunction in pediatric HTx recipients. Methods: We conducted a single center retrospective cohort study on all pediatric HTx recipients (aged < 18) who received a HTx between May 1989 and July 2013 who are currently followed at our center and had BKV testing. Since April 2012, urine BKV loads have been checked in all patients at time of annual evaluation, and prior to that, in patients with chronic kidney disease or acute kidney dysfunction. Serum BK viremia was assayed in patients with positive BK viruria. Data collected included demographics, donor information, immunosuppression regimen, and history of viral illnesses. Descriptive statistics, Fisher’s exact analysis, and Student’s T-test were done to compare patients with and without BK viruria (controls). Results: Of 100 eligible HTx recipients, 51 have been screened to date. Seventeen (33%) of these had BK viruria, and 4 (8%) developed viremia. One viremic patient (2%) progressed to BKV nephropathy diagnosed by biopsy, and required a renal transplant for end stage renal disease. Increasing mean urine viral load was significantly associated with progression from BK viruria (2.1x10^8 copies/mL) to viremia (22.2x10^8 copies/mL) (p= 0.01). History of positive EBV serologies (p= 0.01), use of Sirolimus immunosuppression (p= 0.0006), and duration of time since HTx (6.42 years vs. 3.84 years; p= 0.04) were significantly associated with BK viruria. Renal function, in terms of Schwartz CKiD eGFR, was not significantly different between patients with BK viruria and controls (74.2 vs. 79.8 mL/min per 1.73 m^2, p= 0.41). Conclusion: Increasing mean urine BK viral load, time since HTx, evidence of past EBV infection, and use of Sirolimus were associated with BKV. A prospective study is planned to assess the incidence of BKV in HTx recipients. Routine screening of the urine BK viral load should be considered in this population. 3( 20) Risk Factors for Sudden Death in Children with Cardiac Allograft Vasculopathy B.J. Hong ,1 K. Sexson Tejtel,1 A. Jeewa,1 A.G. Cabrera,1 J.F. Price,1 J.S. Heinle,2 W.J. Dreyer,1 S.W. Denfield.1  1Pediatric Cardiology, Baylor College of Medicine/Texas Children’s Hospital, Houston, TX; 2Pediatric Cardiothoracic Surgery, Baylor College of Medicine/Texas Children’s Hospital, Houston, TX. Purpose: Cardiac allograft vasculopathy (CAV) is one of the leading causes of late mortality in pediatric heart transplant patients. Many do not survive to retransplantation, and a subset die suddenly. The purpose of this study is to determine risk factors associated with sudden death (SD) in pediatric patients with CAV. Methods: This is a single center retrospective review of all pediatric patients transplanted from 1984 to 2012 who died or were retransplanted following a diagnosis of CAV. The diagnosis of CAV was made by coronary angiography, explant pathology, or autopsy. Data including demographics, episodes of rejection, cardiac catheterization and echocardiographic data, time to CAV diagnosis, and cause of death were collected to compare SD patients to those who died of end-stage heart failure. Results: Of the 54 patients with CAV, 40 died or underwent retransplantation (RT) and 14 are alive. Three patients were excluded as their causes of death were unknown or noncardiac. Of the 37 patients who met inclusion criteria, 9 patients (24%) died suddenly. The other 28 patients died of heart failure (12/28 patients) or were retransplanted (16/28 patients). Comparison of the SD group to the heart failure death/RT group did not find any differences in age at transplant, episodes of cellular or antibody mediated rejection, or time to CAV diagnosis or death. Systolic function was significantly better in the SD group with a median ejection fraction of 62% (range 50-70%) compared to 50% (range 24-75%) in the heart failure death/RT group (p= 0.045). Hemodynamics also differed, with a median right ventricular end diastolic pressure of 8 mmHg (range 3-16 mmHg) in the SD group compared to a median of 14 mmHg (range 4-24 mmHg) in the heart failure death/RT group (p= 0.011) and a median pulmonary capillary wedge pressure of 9 mmHg (range 4-16 mmHg) in the SD group compared to a median of 14 mmHg (range 7-27 mmHg) in the heart failure death group/RT (p= 0.027). B-type natriuretic peptide (BNP) levels before death or retransplantation were lower in the SD group with a median of 260 pg/ml (range 81-953 pg/ml) compared