Congenital Heart Disease

Survival Prospects and Circumstances of Death

in Contemporary Adult Congenital Heart Disease Patients
Under Follow-Up at a Large Tertiary Centre
Gerhard-Paul Diller, MD, MSc, PhD, MBA*; Aleksander Kempny, MD*;
Rafael Alonso-Gonzalez, MD, MSc; Lorna Swan, MD, FRCP; Anselm Uebing, MD, PhD;
Wei Li, MD, PhD; Sonya Babu-Narayan MB, BS, BSc, MRCP, PhD;
Stephen J. Wort, PhD; Konstantinos Dimopoulos, MD, MSc, PhD; Michael A. Gatzoulis, MD, PhD

Background—Adult congenital heart disease (ACHD) patients have ongoing morbidity and reduced long-term survival.
Recently, the importance of specialized follow-up at tertiary ACHD centers has been highlighted. We aimed to assess
survival prospects and clarify causes of death in a large cohort of patients at a single, tertiary center.
Methods and Results—We included 6969 adult patients (age 29.9±15.4 years) under follow-up at our institution between
1991 and 2013. Causes of death were ascertained from official death certificates. Survival was compared with the
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018

expected survival in the general age- and sex-matched population, and standardized mortality rates were calculated.
Over a median follow-up time of 9.1 years (interquartile range, 5.2–14.5), 524 patients died. Leading causes of death
were chronic heart failure (42%), pneumonia (10%), sudden-cardiac death (7%), cancer (6%), and hemorrhage (5%),
whereas perioperative mortality was comparatively low. Isolated simple defects exhibited mortality rates similar to those
in the general population, whereas patients with Eisenmenger syndrome, complex congenital heart disease, and Fontan
physiology had much poorer long-term survival (P<0.0001 for all). The probability of cardiac death decreased with
increasing patient’s age, whereas the proportion of patients dying from noncardiac causes, such as cancer, increased.
Conclusions—ACHD patients continue to be afflicted by increased mortality in comparision with the general population
as they grow older. Highest mortality rates were observed among patients with complex ACHD, Fontan physiology,
and Eisenmenger syndrome. Our data provide an overview over causes of mortality and especially the spectrum of
noncardiac causes of death in contemporary ACHD patients.   (Circulation. 2015;132:2118-2125. DOI: 10.1161/
CIRCULATIONAHA.115.017202.)
Key words: heart defects, congenital ◼ heart failure ◼ mortality ◼ sudden cardiac death ◼ survival

L ife expectancy of patients born with congenital heart dis-

ease (CHD) has improved dramatically over the past few
decades.1 In fact, >90% of these patients are now expected
to survive to adulthood.2 This has led to the development
of a large and growing population of adults with congenital
heart disease (ACHD). Despite the surgical, interventional,
and medical advancements, these patients are not cured and
require life-long specialized health care. Beyond the obvi-
ous ongoing morbidity, including cardiac symptoms, reduced
exercise capacity, and the need for electrophysiological,
interventional, or surgical procedures, mortality is increased
in this population of patients with chronic cardiac disease.3
Clinical Perspective on p 2125
Previous studies have investigated the long-term mortality of
various ACHD cohorts and have delineated causes of death
in this population.4–6 Because of ongoing improvement of
care, survival prospects of adults with congenital heart dis-
ease are likely to have changed over recent decades. Khairy
et al7 demonstrated that mortality in patients with congenital
heart disease has shifted away from infants and toward adults,
with a steady increase in age at death. The current study was
designed to evaluate a contemporary ACHD cohort from a
single tertiary center and attempt comparison with data from

Received April 23, 2015; accepted September 8, 2015.

From Adult Congenital Heart Centre and National Centre for Pulmonary Hypertension, Royal Brompton Hospital, London, United Kingdom (G.-P.D.,
A.K., R.A.-G., L.S., A.U., W.L., S.B.-N., S.J.W., K.D., M.A.G.); NIHR Cardiovascular Biomedical Research Unit, Royal Brompton Hospital and
National Heart and Lung Institute, Imperial College London, United Kingdom (G.-P.D., A.K., R.A.-G., L.S., A.U., W.L., S.B.-N., S.J.W., K.D., M.A.G.);
National Heart and Lung Institute, Imperial College School of Medicine, London, United Kingdom (G.-P.D., A.K., R.A.-G., L.S., A.U., W.L., S.B.-N.,
S.J.W., K.D., M.A.G.); and Division of Adult Congenital and Valvular Heart Disease, Department of Cardiology and Angiology, University Hospital
Muenster, Germany (G.-P.D.).
Drs Diller and Kempny contributed equally.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.
115.017202/-/DC1.
Correspondence to Aleksander Kempny, MD, Adult Congenital Heart Centre, Royal Brompton and Harefield NHS Foundation Trust, Sydney Street,
SW3 6NP London, United Kingdom. E-mail a.kempny@rbht.nhs.uk
© 2015 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org DOI: 10.1161/CIRCULATIONAHA.115.017202

2118
 Diller et al   Survival Prospects in Contemporary ACHD Patients   2119

previous studies. In addition, we provide herewith mortality
data in relationship to the general population, adjusted for age
and sex, and propose a novel approach for presenting these
data to health professionals, health policy makers, and patients
alike.
Patients and Methods
We retrospectively reviewed data on all adult patients with congenital
heart disease under active follow-up at the Royal Brompton Hospital,
London between 1991 and 2014. For the scope of this study we
defined the start of adulthood as age ≥16. Patients were divided into
subgroups based on the major underlying heart defect. Patients were
grouped by primary diagnosis into major ACHD groups. Those with
combinations of lesions were assigned to the lesion of highest com-
plexity. Patients with complex cardiac anatomy (comprising mainly
patients with single-ventricle physiology) without Eisenmenger syn-
drome and without Fontan palliation were coded as ‘complex CHD.’
Patients with (generally isolated) congenital valvar lesions were
coded as ‘valvar disease.’ Data on clinical status were obtained from
medical records. Data on overall mortality were retrieved from the
Office for National Statistics, which registers all United Kingdom
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018
competing risk survival model accounting for left truncation. To this
end the cumulative incidence of death in the presence of competing
risk was estimated as described by Coviello and Boggess8 using the
stcompet package for Stata (Version IC 12.1, College Station, TX),
and cumulative incidence plots with 95% confidence intervals were
produced. This statistical approach models simultaneously the risk of
multiple competing outcomes including heart failure, sudden cardiac
death, acute myocardial infarction, cardiac surgery and interventions,
cancer, or pneumonia-related mortality in our study. To estimate
standardized mortality ratios (SMRs) compared with an age- and
sex-matched sample of the general population the method reported
by Finkelstein et al9 was used. Survival was compared with that pre-
dicted for an age- and sex-matched healthy cohort of UK residents
using life table data (2007–2009 interim life tables) published by
the Government Actuary’s Department (http://www.gad.gov.uk), as
previously described.9 Equivalent age was defined as the age of UK
population with the most similar 5-year mortality (ie, minimal sum
of absolute differences). Statistical analyses were performed using
R-package version 3.0.2.10 A 2-sided P-value of <0.05 was consid-
ered indicative of statistical significance.
Results

deaths. The cause of death was established from medical records

and death certificates, available for all patients, by 1 investigator
(G-P.D.). Where the likely immediate cause of mortality remained
unclear, the case was discussed with 1 of the coprincipal investigators
(A.K.) and consensus was reached. In addition, the records of the
deceased patients were cross checked with data from the local surgi-
cal and interventional audit database to ascertain that no perioperative
death was missed. Because this was a retrospective analysis based
on data collected for routine clinical care and administrative pur-
poses (UK National Research Ethics Service guidance), individual
informed consent was not required. The study was locally registered
and approved.
Statistical Analysis
Continuous variables are presented as mean±standard deviation or
median and interquartile range, whereas categorical variables are
presented as number (percentage). The association between vari-
ous causes of mortality and age was assessed using a nonparametric
Demographics and Mortality
We included 6969 patients (49.9% females) under active follow-
up at our institution as illustrated in Table 1. The mean age at
baseline was 29.9±15.4 years. Overall, 69%, 26%, and 5%
of patients were in the New York Health Association (NYHA)
functional class I, II, and III/IV, respectively. According to the
Bethesda disease complexity classification 52% of patients had
simple defects, 33% moderate, and 15% great complexity defects.
During a median follow-up time of 9.1 years (interquartile
range, 5.2–14.5; corresponding to a total of 70 967 patient-
years), 524 (7.7%) patients died yielding a mortality rate of
0.72%/patient-year.
The majority of patients (429; 81.9%) died outside hospi-
tal, whereas the remainder died in our institution or within 24 h
from discharge. Death occurred after an elective or emergency

Table 1.  Demographics and Baseline Characteristics

Age at Baseline, Age at Death, Years
Diagnosis n Female, % Deceased, n (%) Years Median [IQR] NYHA,* % I/II/III/IV
ASD 1,092 61.3 66 (6.2) 39.8±18.3 72 [62–79] 69/28/3/0
PDA 117 77.8 2 (1.7) 32.8±16.4 74 [63–86] 84/11/5/0
VSD 713 50.6 19 (5.9) 26.1±12.5 48 [39–60] 85/12/3/0
Marfan syndrome 296 43.6 23 (7.8) 32.0±14.5 49 [33–60] 93/7/0/0
Valvar disease 1442 44.7 85 (5.9) 29.9±15.6 62 [45–77] 76/20/4/0
Aortic coarctation 860 41.3 39 (4.6) 28.9±14.3 48 [37–65] 88/11/1/0
Ebstein 153 54.9 19 (12.6) 34.5±16.2 52 [37–73] 51/41/8/0
AVSD 255 57.6 15 (5.9) 29.1±14.8 52 [48–63] 70/26/3/0
Tetralogy of Fallot 869 45.9 54 (6.3) 26.8±13.1 50 [35–66] 68/29/3/0
TGA arterial switch 171 30.5 3 (2.4) 17.0±4.4 — 76/21/3/0
Systemic RV 279 46.0 34 (12.5) 27.6±12.4 37 [28–42] 61/30/8/0
Complex CHD 265 52.7 67 (25.8) 24.2±10.1 36 [30–46] 36/55/9/1
Eisenmenger 277 63.5 64 (23.4) 30.6±12.3 42 [33–49] 4/56/38/1
Fontan 180 53.3 34 (19.2) 21.4±7.4 28 [23–36] 43/52/5/0
All patients 6,969 49.9 524 (7.7) 29.9±15.4 47 [34–65] 69/26/5/0
ASD indicates atrial septal defect; AVSD, atrioventricular septal defect; CHD, congenital heart disease; IQR, interquartile range; NYHA,
New York Heart Association Functional Class; PDA, patent ductus arteriosus; RV, right ventricle; TGA, transposition of the great arteries;
and VSD, ventricular septal defect.
 2120  Circulation  December 1, 2015

cardiac operation in 25 patients (Fontan-revision/conversion
related surgery in 6, tricuspid valve surgery in 5, pulmonary
valve replacement in 4, aortic surgery in 4, and other/complex
surgery in 5). In addition, 1 Eisenmenger patient died early
after heart-lung transplantation, whereas 4 patients succumbed
to complications related to cardiac interventional procedures.
Table 2 provides an overview over the causes of death in this
population. It illustrates that the leading cause of mortality in
our cohort was chronic cardiac failure, followed by pneumonia
and sudden cardiac death. Remarkably, cardiac surgery/cardiac
intervention related mortality ranked only 5th in this statistic,
after pneumonia and cancer. The same table also demonstrates
the relatively high proportion of patients dying from noncar-
diac causes such as cancer, hemorrhage (56% cerebral, 19%
pulmonary, 11% gastrointestinal), infection, or cerebrovascular
events. In addition, we provide the percentage of patients dying
because of aortic dissection or hepatic failure, both recognized
causes of mortality in selected subgroups of patients with CHD.
With increasing patient age, the proportion of patients dying
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018
gender matched sample from the general UK population
(SMR, 2.29; 95% CI, 2.08–2.53; Logrank P<0.0001). There
were, however, significant differences in mortality between
subgroups of patients (Logrank P<0.0001; see Figure 3 and
Figure I in the online-only Data Supplement). The SMR was
highest in patients with Fontan circulation (SMR, 23.40; 95%
CI, 15.97–34.29; P<0.0001), complex CHD (SMR, 14.13;
95% CI, 10.71–18.64; P<0.0001), and Eisenmenger syndrome
(SMR, 12.79; 95% CI, 9.67–16.91; P<0.0001). In contrast,
no significant difference in mortality was present in patients
with ductus arteriosus and atrial or ventricular septal defects
when compared with the general UK population (P>0.05, for
all). The SMRs based on the Bethesda classification11 were 1.3
(95% CI, 1.1–1.5), 2.2 (95% CI, 1.8–2.3), and 10.9 (95% CI,
9.3–12.8) for patients with simple, moderate, and great com-
plexity heart defects (P<0.001 for all). In addition, NYHA
functional class was associated with prognosis for the over-
all cohort. The SMR increased from 1.6 (95% CI, 1.3–1.9;
P<0.0001) for class 1, to 3.6 (95% CI, 3.0–4.2; P<0.0001) for

because of cardiac reasons (despite the inclusion of acute myo-
cardial infarction) decreased and, by implication, proportionally
more patients died because of competing noncardiac causes
(Figure 1A). This association was especially evident in patients
with simple defects (Figure 1B). In addition, Figure 2A shows
the association between increasing age and cancer- or pneumo-
nia-related deaths based on the results of a competing risk sur-
vival analysis. Regarding reasons for cardiac death, especially
the risk of heart failure–related death and that of acute myocar-
dial infarction–related mortality, increased with age (Figure 2B).
Survival in the entire ACHD cohort was significantly
worse compared with the expected mortality for an age and
class 2 and 4.6 (95% CI, 3.6–6.0; P<0.0001) for class 3 or 4.
Based on the fitted SMR models we calculated pre-
dicted 5-year risk of death for each diagnostic subgroup for
hypothetical 40-year-old patients with CHD. These mortal-
ity risks were compared with the projected risk of the gen-
eral population to obtain an ‘equivalent age’ with regards
to mortality risk for each ACHD subgroup (Figure 4). For
example, a 40-year-old average patient with Fontan physi-
ology from our cohort had a 5-year risk of death (18.0%;
95% CI, 11.9–24.6%) comparable with that of a 75-year-
old person without CHD. In addition, Figure 5 illustrates
the equivalent age for the different diagnostic groups and

Table 2.  Distribution of Causes of Death in Different Diagnostic Subgroups and Overall
TGA
Aortic Complex Marfan Systemic Tetralogy arterial Valvular All
Rank Cause of death coarctation ASD AVSD CHD Ebstein Eisenmenger Fontan syndrome PDA RV of Fallot switch disease VSD patients
1 Heart failure 31% 28% 57% 57% 38% 45% 52% 30% 50% 66% 40% — 40% 39% 42.5%
2 Pneumonia 8% 17% 7% 2% — 16% — 5% — — 18% — 17% 6% 10.2%
3 Sudden cardiac — — 14% 11% — 9% 13% 5% — 13% 6% 33% 6% 11% 7.0%
death
4 Cancer 11% 14% 7% — 13% — 3% 25% 50% — 4% — 2% 22% 6.3%
5 Cardiac Surgery/ 11% 2% 7% 5% 19% 2% 19% 5% — 9% 12% — 1% — 6.1%
Intervention
(Cardiac surgery, (5%) (1%) (7%) (1%) (17%) (2%) (18%) (5%) (—) (6%) (11%) (—) (1%) (—) (4.8%)
peri-op)
6 Hemorrhage 8% 6% — 8% — 9% 3% 5% — — — — 7% 11% 5.5%
7 Sepsis/infection — 8% — 6% 6% 3% 3% — — — 8% — 4% 11% 4.5%
8 Cerebrovascular 8% 8% 7% — 6% 5% — — — 3% — — 7% — 4.1%
9 Acute myocardial 6% 5% — 3% — — — — — 3% 8% 33% — — 2.7%
infarction
10 Endocarditis — — — 2% 6% 2% — 10% — — 4% — 1% — 1.6%
.. Aortic dissection 11% 2% — — — — — 5% — — — — 1% — 1.4%
.. Hepatic failure — 5% — — 6% — 3% — — 3% — — 1% — 1.4%
only 2 only 3
deaths deaths
The ‘Cardiac-Surgery/peri-op’ row provides information on mortality related specifically to adult congenital heart surgery. ASD indicates atrial septal defect; AVSD,
atrioventricular septal defect; CHD, congenital heart disease; PDA, patent ductus arteriosus; RV, right ventricle; TGA, transposition of the great arteries; and VSD,
ventricular septal defect.
 Diller et al   Survival Prospects in Contemporary ACHD Patients   2121
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018

Figure 1. Cumulative incidence of cardiac and noncardiac death with 95% confidence intervals (dotted line) for (A) the entire study cohort
and (B) stratified by disease complexity according to the Bethesda classification based on the results of competing risk model. CHD
indicates congenital heart disease.

various ages in comparison to that observed in persons with-
out congenital heart disease.
Discussion
Our data provide a contemporary overview over the causes of
mortality in ACHD patients followed at a large, established
supraregional center. In comparison with previous reports a
shift from perioperative death related to ACHD surgery to
long-term cardiac and especially noncardiac mortality was
evident. Moreover, long-term survival prospects of patients
with simple, isolated congenital defects were found to be
excellent and not statistically different from those expected
 2122  Circulation  December 1, 2015
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018
in the general UK population. The slightly elevated overall
risk of death in patients with simple defects, as a group, is
likely related to the inclusion of patients with valvar defects
in this group. In contrast, patients with uncorrected, palliated,
complex, or cyanotic underlying heart defects continue to be
afflicted by substantial mortality. In addition, mortality rates
in various diagnostic subgroups were compared with the mor-
tality observed in the general population. To illustrate survival
prospects, we introduce the concept of equivalent age. We
contend that this may aid counseling of patients by projecting
mortality risks for individual diagnostic subgroups compared
to what is naturally expected at older age.
Previous studies have investigated primary causes of mor-
tality in ACHD patients; these studies were different from the
present report either because they referred to historical ACHD
cohorts or because they represented registry studies includ-
ing patients followed-up at numerous institutions. Oechslin
Figure 2. A, Cumulative incidence of
pneumonia and cancer death with 95%
confidence intervals based on the results
of competing risk model. B, Cumulative
incidence of various causes of cardiac
mortality with 95% confidence intervals
based on the results of competing risk
model. AMI indicates acute myocardial
infarction.
and Connolly4,12 have described the circumstances of death
in ACHD patients under follow-up at 2 large supraregional
Canadian and US centers (Toronto and Mayo Clinic) in the
1980s and early 1990s, respectively. They reported a periop-
erative mortality of 18% and 37.7%, in what are now historic
cohorts. These mortality rates were largely consistent with the
proportion of patients dying perioperatively (26.3%) reported
by Nieminen et al13 as part of a population-based Finnish
study (albeit the latter study included also initial repair in chil-
dren). In contrast, our data suggest that the focus of ACHD
mortality has nowadays shifted to long-term cardiac and non-
cardiac complications of the disease. Moreover, the periopera-
tive ACHD mortality reported here is even lower compared
with results from a recent Dutch national registry (7.1% peri-
operative deaths between 2002 and 2008),6 supporting the
role of concentrating care at tertiary ACHD centres.14 The
proportion of patients dying from heart failure in our study
 Diller et al   Survival Prospects in Contemporary ACHD Patients   2123

Figure 3. Standardized mortality ratios (SMR) in

various subgroups of patients. Points present the
SMR, and horizontal lines the 95% confidence
interval range. An SMR of 1 suggests that patients
have comparable mortality as a sex- and age-
matched sample from the general population. ASD
indicates atrial septal defect; AVSD, atrioventricular
septal defect; CHD, congenital heart disease;
PDA, patent ductus arteriosus; RV, right ventricle;
TGA, transposition of the great arteries; and VSD,
ventricular septal defect.
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018

is, however, similar to that seen in previous studies.4,5,13 It is Discussing life expectancy issues and short- to midterm
likely that frequency of heart failure is increasing in ACHD risk of death with patients can be challenging. Beyond, obvi-
patients,15 and—given the increasing complexity of disease as ous psychological barriers and anxiety associated with this
well as the growing incidence of comorbid conditions—more
patients present with advanced forms of heart failure. On the
other hand, progress in the management of advanced heart
failure in ACHD has been slow and arguably unsatisfactory.
The fact remains that standard heart failure therapy has still
an unproven and possibly limited effect in this heterogeneous
group of patients,16–18 whereas novel therapeutic options such
as cardiac resynchronization therapy and assist systems have
had a limited uptake so far. In contrast, sudden cardiac death
rate was lower in the present study compared with previous
reports, probably as a result of better risk stratification19–21 and
more liberal use of implantable cardiac defibrillators in the
current era.7 The most remarkable finding, however, was the
relatively large proportion of patients dying due to noncar-
diac complications, including cancer, cerebrovascular disease,
infection, and pneumonia. This is consistent with previous
data published by Khairy et al,7 Afilalo et al,22 and our group.23
Presumably as a consequence of the aging ACHD population
the main causes of mortality are changing. Similar to these
previous studies we could confirm that, with increasing age,
the proportion of patients succumbing to myocardial infarc-
tion increases. However, we could not confirm that acute
myocardial infraction was a leading cause of death, either in
noncyanotic patients as a whole, or in a specific subgroup of
patients. This is in contrast to a population-based US study,
reporting acute myocardial infraction as the leading cause of
death in elderly noncyanotic ACHD patients.24
It is not surprising that survival prospects of ACHD
Figure 4. Projected 5-year mortality rates for 40-year-old ACHD
patients are inferior to those observed in the general popula- patients compared with that expected for the general UK
tion. However, Figure 3 illustrates that especially Fontan, population based on the results of the standardized mortality
Eisenmenger syndrome, and complex CHD patients have ratio (SMR) analysis. Points present the estimated mortality
within 5 years (on the x axis) and also indicate the equivalent
greatly increased mortality rates. In contrast, simple defects age—expressed as the age of subgroup of UK population with
were not found to fare significantly worse in terms of survival the most similar 5-years mortality (y axis). Red lines represent
compared to the general population. We believe our findings 95% confidence intervals for the 5-year mortality. The black
are a testimony to the advance in the CHD field, represent the curve presents 5-year mortality for the UK-population based
on life table data. ASD indicates atrial septal defect; AVSD,
challenges ahead and may help to identify subgroups of patients atrioventricular septal defect; RV, right ventricle; and VSD,
where current and future research efforts need to be intensified. ventricular septal defect.
 2124  Circulation  December 1, 2015
Figure 5. Mortality in subgroups of patients compared with
mortality in age-matched UK population. Numbers on the colored
surface present the equivalent age—expressed as the age of
subgroup of UK population, having similar 5-year mortality rates.
Colors reflect the difference between the relative age and the
actual age of patients. The curved line corresponds to the 5-year
risk of death of the general UK population according to life table
data. ASD indicates atrial septal defect; AVSD, atrioventricular
septal defect; CHD, congenital heart disease; RV, right ventricle;
and VSD, ventricular septal defect.
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018
difficult subject, there may also be a cognitive problem in
understanding risks. One of the key deliverables of the current
report is, therefore, the data on equivalent ages. The mortal-
ity of ACHD patients can be illustrated by comparing it with
that observed in people without heart disease. Although the
equivalent ages for simple lesions are comparable or only
slightly (and nonsignificantly) elevated compared with those
of the general public, patients with complex, univentricular, or
uncorrected lesions had highly elevated values. For example,
an average Fontan 40-year-old patient in our study had mor-
tality rates comparable with those expected in a 75-year-old
individual. We contend that, unlike SMRs and Cox regres-
sion-model derived hazard ratios, these numbers are useful
for counseling patients because they should be intuitively
clear to anybody. Furthermore, unlike percentages or ratios
of mortality equivalent ages implicitly express the stochastic
nature of such estimates. Beyond, direct medical applications,
these data could also be useful for example for (life) insurance
purposes.
Strength of the Current Report
To the best of our knowledge, the current report represents
the largest single center study assessing the causes of mor-
tality in contemporary patients (70 967 patient-years versus
25 900 patient years in a previous nationwide registry study).
This is explained by the relatively long history and the well-
established nature of our center. Compared with a previously
published national registry database (6933 patients, 197
deceased)5 and a pan-European registry study (4110 patients,
115 deceased),25 a group of 524 deceased patients formed the
statistical basis of the current report. In addition, unlike reg-
istry data we had access to the entire medical/surgical data-
base of the patients and could clarify equivocal information
based on original medical records. This approach has been
described to improve data quality and reliability of mortality
data in the setting of ACHD.26 The mortality data presented
here are based on official death certificates complimented by
additional information available to us and should, therefore,
provide robust estimates of the causes of mortality. A further
theoretical advantage of this single center study is the consis-
tent approach with a shared diagnostic and therapeutic strat-
egy used over time.
Limitations
Because this represents a single-center retrospective study, the
sample of patients included may not necessarily represent the
pattern of ACHD patients present in the community. Studying
long-term outcomes of community based ACHD patients is,
therefore, a recognized strength of registry based studies. Like
all similar studies, the distinction between the primary cause
of death is not always unequivocal (eg, pneumonia, which
may be a consequence of cardiac pulmonary congestion).
However, all causes of deaths were checked for plausibility
through comparison with our clinical database and especially
the data on surgical mortality is cross validated with informa-
tion from our clinical/official surgical audit to improve data
quality and minimize the number of patients with death due to
unspecified reasons. The proportion of patients dying periop-
eratively is not equivalent to surgical mortality. Formally, the
former is a function of surgical mortality, competing risks of
deaths, and the number of operations performed. Therefore,
this parameter cannot be compared directly with other stud-
ies reporting specifically surgical mortality rates. However, it
can be compared with previous studies investigating circum-
stances of death in ACHD patients, in general, using the same
metric.4,5,13 Estimates of mortality provided are for average
patients stratified by diagnosis. Therefore, individual patients
may exhibit different mortality, depending on additional fac-
tors specific to the patient. Furthermore, the average age and
median age at death was different between the diagnostic
groups. Especially for the young patients with arterial switch
TGA it may be premature to extrapolate to mortality risks at
older age and these data should, therefore, interpreted with
caution.
Conclusions
The current report confirms that ACHD patients continue to
be afflicted by increased mortality compared with general
population as they grow older. Highest mortality rates were
observed among patients with complex ACHD, Fontan physi-
ology, and Eisenmenger syndrome. Our contemporary data
show a clear shift from perioperative to chronic cardiac mor-
tality and noncardiac death.
Sources of Funding
Dr Kempny was supported by the Deutsche Herzstiftung e.V. Prof
Gatzoulis and the Adult Congenital Heart Center and National Center
for Pulmonary Hypertension have received support from the Clinical
Research Committee and the British Heart Foundation. This project
was supported by the National Institute of Health Research cardiovas-
cular Biomedical Research Unit at the Royal Brompton and Harefield
National Health Service Foundation Trust and Imperial College
London. Sonya V. Babu-Narayan is supported by an Intermediate
Clinical Research Fellowship from the British Heart Foundation
(FS/11/38/28864).
Disclosures
None.
 Diller et al   Survival Prospects in Contemporary ACHD Patients   2125

References 15. Zomer AC, Vaartjes I, van der Velde ET, de Jong HM, Konings TC,
Wagenaar LJ, Heesen WF, Eerens F, Baur LH, Grobbee DE, Mulder BJ.
1. Gatzoulis MA. Adult congenital heart disease: education, education,
Heart failure admissions in adults with congenital heart disease; risk fac-
education. Nat Clin Pract Cardiovasc Med. 2006;3:2–3. doi: 10.1038/
tors and prognosis. Int J Cardiol. 2013;168:2487–2493. doi: 10.1016/j.
ncpcardio0382.
ijcard.2013.03.003.
2. Moons P, Bovijn L, Budts W, Belmans A, Gewillig M. Temporal trends in
16. Bolger AP, Coats AJ, Gatzoulis MA. Congenital heart disease: the original
survival to adulthood among patients born with congenital heart disease
heart failure syndrome. Eur Heart J. 2003;24:970–976.
from 1970 to 1992 in Belgium. Circulation. 2010;122:2264–2272. doi:
17. Bolger AP, Gatzoulis MA. Towards defining heart failure in adults with
10.1161/CIRCULATIONAHA.110.946343.
congenital heart disease. Int J Cardiol. 2004;97 Suppl 1:15–23. doi:
3. Baumgartner H, Bonhoeffer P, De Groot NM, de Haan F, Deanfield JE,
10.1016/j.ijcard.2004.08.005.
Galie N, Gatzoulis MA, Gohlke-Baerwolf C, Kaemmerer H, Kilner
18. van der Bom T, Winter MM, Bouma BJ, Groenink M, Vliegen HW, Pieper
P, Meijboom F, Mulder BJ, Oechslin E, Oliver JM, Serraf A, Szatmari
PG, van Dijk AP, Sieswerda GT, Roos-Hesselink JW, Zwinderman AH,
A, Thaulow E, Vouhe PR, Walma E; Task Force on the Management
Mulder BJ. Effect of valsartan on systemic right ventricular function: a
of Grown-up Congenital Heart Disease of the European Society of
double-blind, randomized, placebo-controlled pilot trial. Circulation.
Cardiology (ESC); Association for European Paediatric Cardiology
2013;127:322–330. doi: 10.1161/CIRCULATIONAHA.112.135392.
(AEPC); ESC Committee for Practice Guidelines (CPG). ESC Guidelines
19. Babu-Narayan SV, Kilner PJ, Li W, Moon JC, Goktekin O, Davlouros
for the management of grown-up congenital heart disease (new version
PA, Khan M, Ho SY, Pennell DJ, Gatzoulis MA. Ventricular fibro-
2010). Eur Heart J. 2010;31:2915–2957. doi: 10.1093/eurheartj/ehq249.
sis suggested by cardiovascular magnetic resonance in adults with
4. Oechslin EN, Harrison DA, Connelly MS, Webb GD, Siu SC. Mode
repaired tetralogy of fallot and its relationship to adverse markers
of death in adults with congenital heart disease. Am J Cardiol.
of clinical outcome. Circulation. 2006;113:405–413. doi: 10.1161/
2000;86:1111–1116.
CIRCULATIONAHA.105.548727.
5. Verheugt CL, Uiterwaal CS, van der Velde ET, Meijboom FJ, Pieper PG,
20. Gatzoulis MA, Balaji S, Webber SA, Siu SC, Hokanson JS, Poile C,
van Dijk AP, Vliegen HW, Grobbee DE, Mulder BJ. Mortality in adult
Rosenthal M, Nakazawa M, Moller JH, Gillette PC, Webb GD, Redington
congenital heart disease. Eur Heart J. 2010;31:1220–1229. doi: 10.1093/
AN. Risk factors for arrhythmia and sudden cardiac death late after repair
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018

eurheartj/ehq032.
of tetralogy of Fallot: a multicentre study. Lancet. 2000;356:975–981. doi:
6. Zomer AC, Vaartjes I, Uiterwaal CS, van der Velde ET, van den Merkhof
10.1016/S0140-6736(00)02714-8.
LF, Baur LH, Ansink TJ, Cozijnsen L, Pieper PG, Meijboom FJ, Grobbee
21. Khairy P, Harris L, Landzberg MJ, Viswanathan S, Barlow A, Gatzoulis
DE, Mulder BJ. Circumstances of death in adult congenital heart disease.
MA, Fernandes SM, Beauchesne L, Therrien J, Chetaille P, Gordon E,
Int J Cardiol. 2012;154:168–172. doi: 10.1016/j.ijcard.2010.09.015.
Vonder Muhll I, Cecchin F. Implantable cardioverter-defibrillators
7. Khairy P, Ionescu-Ittu R, Mackie AS, Abrahamowicz M, Pilote L, Marelli
in tetralogy of Fallot. Circulation. 2008;117:363–370. doi: 10.1161/
AJ. Changing mortality in congenital heart disease. J Am Coll Cardiol.
CIRCULATIONAHA.107.726372.
2010;56:1149–1157. doi: 10.1016/j.jacc.2010.03.085.
22. Afilalo J, Therrien J, Pilote L, Ionescu-Ittu R, Martucci G, Marelli AJ.
8. Coviello V, Boggess M. Cumulative incidence estimation in the presence
Geriatric congenital heart disease: burden of disease and predictors
of competing risks. Stata J. 2004;4:103–112.
of mortality. J Am Coll Cardiol. 2011;58:1509–1515. doi: 10.1016/j.
9. Finkelstein DM, Muzikansky A, Schoenfeld DA. Comparing sur-
jacc.2011.06.041.
vival of a sample to that of a standard population. J Natl Cancer Inst.
23. Tutarel O, Kempny A, Alonso-Gonzalez R, Jabbour R, Li W, Uebing
2003;95:1434–1439.
A, Dimopoulos K, Swan L, Gatzoulis MA, Diller GP. Congenital heart
10. R developement Core Team. R: A language and environment for statis-
disease beyond the age of 60: emergence of a new population with high
tical computing. Vienna, austria: R foundation for statistical computing.
resource utilization, high morbidity, and high mortality. Eur Heart J.
Retrieved from http://www.R-project.Org. 2010.
2014;35:725–732. doi: 10.1093/eurheartj/eht257.
11. Warnes CA, Liberthson R, Danielson GK, Dore A, Harris L, Hoffman
24. Pillutla P, Shetty KD, Foster E. Mortality associated with adult congenital
JI, Somerville J, Williams RG, Webb GD. Task force 1: the chang-
heart disease: Trends in the US population from 1979 to 2005. Am Heart
ing profile of congenital heart disease in adult life. J Am Coll Cardiol.
J. 2009;158:874–879. doi: 10.1016/j.ahj.2009.08.014.
2001;37:1170–1175.
25. Engelfriet P, Boersma E, Oechslin E, Tijssen J, Gatzoulis MA, Thilén U,
12. Connolly H, Ammash N, Warnes C. One thousand patients with adult
Kaemmerer H, Moons P, Meijboom F, Popelová J, Laforest V, Hirsch R,
congenital heart disease: Causes of death and lessons learned. J Am Coll
Daliento L, Thaulow E, Mulder B. The spectrum of adult congenital heart
Cardiol. 1997;29A:385A–385A.
disease in Europe: morbidity and mortality in a 5 year follow-up period.
13. Nieminen HP, Jokinen EV, Sairanen HI. Causes of late deaths after

The Euro Heart Survey on adult congenital heart disease. Eur Heart J.
pediatric cardiac surgery: a population-based study. J Am Coll Cardiol.
2005;26:2325–2333. doi: 10.1093/eurheartj/ehi396.
2007;50:1263–1271. doi: 10.1016/j.jacc.2007.05.040.
26. Zomer AC, Uiterwaal CS, van der Velde ET, Tijssen JG, Mariman EC,
14. Mylotte D, Pilote L, Ionescu-Ittu R, Abrahamowicz M, Khairy P, Therrien
Verheugt CL, Vaartjes I, Pieper PG, Meijboom FJ, Grobbee DE, Mulder
J, Mackie AS, Marelli A. Specialized adult congenital heart disease care:
BJ. Mortality in adult congenital heart disease: are national registries reli-
the impact of policy on mortality. Circulation. 2014;129:1804–1812. doi:
able for cause of death? Int J Cardiol. 2011;152:212–217. doi: 10.1016/j.
10.1161/CIRCULATIONAHA.113.005817.
ijcard.2010.07.018.

Clinical Perspective
Adult congenital heart disease (ACHD) patients have ongoing morbidity and reduced long-term survival. Based on a large
sample of patients under follow-up at an established supraregional ACHD center we provide data on contemporary survival
prospects and clarify causes of death in this challenging population. The current analysis includes 6969 adult patients (mean
age 30 years) under follow-up at our institution between 1991 and 2013. Causes of death were ascertained from official death
certificates. Survival was compared with the expected survival in the general age- and sex-matched population, and standard-
ized mortality rates were calculated. Leading causes of death were chronic heart failure, pneumonia, sudden-cardiac death,
and cancer, whereas perioperative ACHD mortality was comparatively low. Our data illustrate that ACHD patients continue
to be afflicted by increased mortality compared to the general population as they grow older. Highest mortality rates were
observed among patients with complex congenital heart disease, Fontan physiology, and Eisenmenger syndrome. Our data
also show a shift from perioperative to chronic cardiac mortality and noncardiac death. In addition, we introduce the concept
of equivalent age, which may aid counseling of patients by projecting mortality risks for individual diagnostic subgroups
compared to what is naturally expected at older age.
 Survival Prospects and Circumstances of Death in Contemporary Adult Congenital Heart
Disease Patients Under Follow-Up at a Large Tertiary Centre
Gerhard-Paul Diller, Aleksander Kempny, Rafael Alonso-Gonzalez, Lorna Swan, Anselm
Uebing, Wei Li, Sonya Babu-Narayan, Stephen J. Wort, Konstantinos Dimopoulos and Michael
A. Gatzoulis
Downloaded from http://circ.ahajournals.org/ by guest on March 15, 2018

Circulation. 2015;132:2118-2125; originally published online September 14, 2015;

doi: 10.1161/CIRCULATIONAHA.115.017202
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2015 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/132/22/2118

Data Supplement (unedited) at:

http://circ.ahajournals.org/content/suppl/2015/09/14/CIRCULATIONAHA.115.017202.DC1

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Circulation is online at:

http://circ.ahajournals.org//subscriptions/
 SUPPLEMENTAL MATERIAL

Supplemental Figure 1

Kaplan Meier survival curves compared to expected mortality of an age and gender matched sample from the general UK population stratified by
diagnostic group. SMR = standardized mortality rates.