Accepted Manuscript

Title: Mortality Risk Stratification in Small Patient Cohorts; the Heart

Transplant Post Fontan Paradigm

Author: Stavros Polyviou, John O'Sullivan, Asif Hasan, Louise Coats

PII: S0002-9149(18)30389-8
DOI: https://doi.org/10.1016/j.amjcard.2018.03.021
Reference: AJC 23213

To appear in: The American Journal of Cardiology

Received date: 23-2-2018

Please cite this article as: Stavros Polyviou, John O'Sullivan, Asif Hasan, Louise Coats,
Mortality Risk Stratification in Small Patient Cohorts; the Heart Transplant Post Fontan Paradigm,
The American Journal of Cardiology (2018), https://doi.org/10.1016/j.amjcard.2018.03.021.

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 proof before it is published in its
final 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.
Title:

Mortality risk stratification in small patient cohorts; the heart transplant post Fontan

paradigm

Authors:

Stavros Polyviou, MD, PhDa, John O'Sullivan, MDa,b, Asif Hasan, FRCS C/Tha,

Louise Coats, PhDa,b

a
Adult Congenital and Paediatric Heart Unit, Freeman Hospital, Newcastle upon

Tyne, UK

b
Cardiovascular Research Centre, Institute of Genetic Medicine, Newcastle

University, Newcastle upon Tyne, UK

Corresponding author: stavrospolyviou@gmail.com (Stavros Polyviou).

Text:

We read with interest the recently published mortality risk stratification for

Fontan patients undergoing heart transplantation by Berg et al1. We would like to

congratulate the authors on the comprehensive nature of their work and

acknowledge that such a tool could have great utility in informing the difficult decision

making around listing this high risk group for cardiac transplantation.

To assess the applicability of their findings to other cohorts, we applied the

risk scoring factors used by Berg et al to all patients who underwent heart

transplantation for failing Fontan between January 2009 and October 2017 at our

institution (we excluded cases prior to 2009 as this phase may be considered to

represent a learning curve). One patient who underwent combined heart and liver

transplant was also excluded2. Points were allocated according to the pre-transplant

risk factors, as suggested (1 point: age<18 years, 1 point: systemic ventricular

Page 1 of 5
ejection fraction<20%, 2 points: Fontan-to-transplant interval <10 years, 2 points:

MELD-XI ≥19, 3 points: moderate systemic atrioventricular valve regurgitation, 3

points: need of preoperative advanced cardiac or renal mechanical support). We

then assessed mortality rates according to the three suggested risk groups (Group 1:

score 0-1, Group 2: score 2-5 and Group 3: score 6-12 points). The association of

all-cause mortality with individual demographic, echocardiographic and clinical

variables was then assessed using the Cox proportional hazards model in our group.

Thirty-seven patients (20/37 male, median age 21.3 years [range 3.4 - 43.5],

with a median follow-up of 2.8 years [range 0-8.1 years]) were included. In our

cohort, Group 1 had a mortality of 33.3%, Group 2 had a mortality of 43.8% and

Group 3 had a mortality of 16.7% respectively. Most deaths (10/13) were early and

preceded hospital discharge. All paediatric patients (<18 years, n=14, 5 deaths) had

a Fontan-to-transplant interval <10 years and 21/23 adult patients (>18 years, 8

deaths) had an interval >10 years, revealing an expected significant overlap between

these two risk factors. Risk factors identified in our population are presented in the

Table and contrast with the findings of Berg et al.

Composition of patient populations and different clinical thresholds for

transplantation will influence the identification of risk factors when single centre

patient cohorts are interrogated. In our experience, the nature of Fontan failure

differs between the paediatric and adult age groups. In our recently published adult

cohort, for example, preserved ventricular function conferred risk, presumably as it

tends to reflect advanced physiology and multisystem disease which is associated

with a more difficult post-transplant course3. Further patients in extremis at the time

of transplantation will be at higher risk of early post-operative death but the causes

and risk factors for late death are likely to be different. Development and validation

of robust risk scores for such a heterogenous group of patients will require large

Page 2 of 5
numbers to permit clarification of the questions being asked and therefore scores

derived from small cohorts, should be treated with caution.

Ideally, the component parts of a risk score should be independent factors to

avoid skewing results. Interdependence, in the score presented by Berg et al.

between age and Fontan-to-transplant interval, should be highlighted and is

confirmed in our data, but interdependence may also exist between MELD-XI

(derived from bilirubin and creatinine) and the need for renal replacement therapy.

Additionally, clarity is needed as to when serum markers are measured in

relationship to transplantation, as some centres, including ours, often optimise end-

organ function prior to transplantation with intravenous inotropes and diuresis2.

In conclusion, cardiac transplantation for the failing Fontan patient remains a

high-risk enterprise and a risk score to guide decision making would be valuable. The

results of the study by Berg et al alongside our findings should provide stimulus for

multi-centre co-operation to try and identify patients with risks that may be

considered prohibitive.

References:

1. Berg CJ, Bauer BS, Hageman A, Aboulhosn JA, Reardon LC. Mortality Risk

Stratification in Fontan Patients Who Underwent Heart Transplantation. Am J Cardiol

2017;119:1675–1679. Available at:

http://linkinghub.elsevier.com/retrieve/pii/S0002914917301911.

2. Duong P, Coats L, O’Sullivan J, Crossland D, Haugk B, Babu-Narayan S V.,

Keegan J, Hudson M, Parry G, Manas D, Hasan A. Combined heart-liver

transplantation for failing Fontan circulation in a late survivor with single-ventricle

physiology. ESC Hear Fail 2017:675–678. Available at:

http://doi.wiley.com/10.1002/ehf2.12202.

3. Murtuza B, Hermuzi A, Crossland DS, Parry G, Lord S, Hudson M, Chaudhari MP,

Haynes S, O’Sullivan JJ, Hasan A. Impact of mode of failure and end-organ

Page 3 of 5
dysfunction on the survival of adult Fontan patients undergoing cardiac

transplantation. Eur J Cardiothorac Surg 2017;51:135–141.

Page 4 of 5
Table 1. Risk factors for transplantation in Fontan patients. The pre-
transplantation risk factors studied, the hazard ratios for death and the 95%
confidence intervals are presented. P values suggesting statistical significance
(P<0.05) are presented in bold.

Variable Hazard Ratio (95% CI) P value

Male gender 0.697 0.524

(0.23-2.12)

Age at Fontan 1.001 0.824

(0.993-1.009)

Moderate/severe systemic AVVR 0.88 0.822

(0.3-2.63)

Preserved systemic ventricular 4.53 0.007

function
(1.5-13.66)

MELD-XI 1.14 0.025

(1.02-1.27)

Renal replacement therapy 4.9 0.018

(1.32-18.22)

Pre-transplantation ECMO 0.046 0.634

(0-14107)

Fontan-transplantation interval 0.999 0.696

(0.994-1.004)

Transplantation age 0.995 0.825

(0.95-1.04)

Page 5 of 5