117
ORIGINAL ARTICLES Increasing Mortality Burden among Adults with Complex Congenital Heart Disease Matthias Greutmann, MD,*†1 Daniel Tobler, MD,†‡1 Adrienne H. Kovacs, PhD,† Mehtap Greutmann-Yantiri, MD,† Sarah R. Haile, PhD,§¶ Leonhard Held, PhD,§ Joan Ivanov, MSc, PhD,†** William G. Williams, MD,†† Erwin N. Oechslin, MD,† Candice K. Silversides, MD,† and Jack M. Colman, MD† *Adult Congenital Heart Disease Program, University Hospital Zurich, Zurich, Switzerland; †Toronto Congenital Cardiac Centre for Adults, Peter Munk Cardiac Centre, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada; ‡Adult Congenital Heart Disease Clinic, University Hospital Basel, Basel, Switzerland; § Division of Biostatistics, Institute for Social and Preventive Medicine, University of Zurich, Zurich, Switzerland; ¶Clinical Trials Unit, Kantonsspital St. Gallen, St. Gallen, Switzerland; **Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada; ††Department of Cardiovascular Surgery, Hospital for Sick Children, Toronto, Ontario, Canada ABSTRACT
Background. Progress in management of congenital heart disease has shifted mortality largely to adulthood. However, adult survivors with complex congenital heart disease are not cured and remain at risk of premature death as young adults. Thus, our aim was to describe the evolution and mortality risk of adult patient cohorts with complex congenital heart disease. Methods. Among 12 644 adults with congenital heart disease followed at a single center from 1980 to 2009, 176 had Eisenmenger syndrome, 76 had unrepaired cyanotic defects, 221 had atrial switch operations for transposition of the great arteries, 158 had congenitally corrected transposition of the great arteries, 227 had Fontan palliation, and 789 had repaired tetralogy of Fallot. We depict the 30-year evolution of these 6 patient cohorts, analyze survival probabilities in adulthood, and predict future number of deaths through 2029. Results. Since 1980, there has been a steady increase in numbers of patients followed, except in cohorts with Eisenmenger syndrome and unrepaired cyanotic defects. Between 1980 and 2009, 308 patients in the study cohorts (19%) died. At the end of 2009, 85% of survivors were younger than 50 years. Survival estimates for all cohorts were markedly lower than for the general population, with important differences between cohorts. Over the upcoming two decades, we predict a substantial increase in numbers of deaths among young adults with subaortic right ventricles, Fontan palliation, and repaired tetralogy of Fallot. Conclusions. Anticipatory action is needed to prepare clinical services for increasing numbers of young adults at risk of dying from complex congenital heart disease. Key Words. Adult Congenital Heart Disease; Congenital Heart Surgery; Mortality; Outcomes
Introduction
C
omplex congenital heart defects affect about 0.6% of newborns, with stable incidence over time.1,2 Most affected patients are now expected to survive to adulthood.3,4 However, adults with repaired complex congenital heart lesions and with 1
These authors contributed equally to the study design, data interpretation, and manuscript preparation.
unrepaired cyanotic lesions (including large shunt lesions complicated by Eisenmenger syndrome) remain at risk for long-term complications and are at increased risk for death as young adults.5–8 Surgical repair techniques for complex congenital heart lesions were introduced several decades ago and have subsequently undergone continuous modification. Thus, cohorts of adult survivors are still emerging and evolving in terms
Congenit Heart Dis. 2015;10:117–127 © 2014 The Authors. Congenital Heart Disease Published by Wiley Periodicals, Inc. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
118 of numbers, age distribution, complexity, previous interventional management, and survival patterns. In previous analyses of deaths in adults with congenital heart disease, cohorts with Eisenmenger syndrome, unrepaired cyanotic heart defects, atrial switch operations for transposition of the great arteries, congenitally corrected transposition of the great arteries, Fontan palliation for univentricular physiology, and repaired tetralogy of Fallot contributed disproportionately to the number of young adults dying from their disease.5,6,8,9 Understanding the evolution of such adult patient cohorts and their mortality risk is essential for planning future care and research. The aims of this study were (1) to examine the 30-year evolution of 6 cohorts of adult survivors of common complex congenital heart lesions at a large tertiary care center for adults with congenital heart disease; (2) to analyze survival probabilities in these patient cohorts; and (3) to predict future trends in numbers of deaths over the upcoming two decades.
Methods
Database and Patient Cohorts The Toronto Congenital Cardiac Centre for Adults at Toronto General Hospital was founded in 1959. From its inception, it has collaborated closely with the Toronto Hospital for Sick Children (SickKids), which is the largest pediatric cardiology center in Canada. At SickKids, congenital cardiac surgery began in 1944, and the first open-heart operation using a heart–lung machine took place in 1958. SickKids surgeons have performed 80–90% of all congenital cardiac surgical procedures for the province of Ontario, Canada. Over the study period, the population in Ontario increased from 9 million to 13 million in 2009. The clinical and administrative database at the Toronto Congenital Cardiac Centre for Adults includes all adults with congenital heart disease ever followed by the program, whether strictly ambulatory or as inpatients or both. It contains information about the anatomical cardiac diagnoses and previous surgical and nonsurgical treatment history. As well, the database is used to track follow-up visits. If patients do not attend scheduled follow-up visits they and, as necessary, their alternate contacts and general practitioners are contacted by administrative staff. Survival status is regularly updated. The mode of death of patients Congenit Heart Dis. 2015;10:117–127
Greutmann et al. who died at our institution up to 1996 has previously been reported.6 In the past, survival of children with complex congenital heart disease was critically dependent on the availability of surgical repair at a given center. In the early days of surgical repair, improving perioperative care led to increased survival of affected children. In contrast, long-term survival of those who survived the operation and early complications, reaching adulthood, was not further impacted by era of birth cohort.10–12 Indeed, a recent study from Belgium demonstrated that survival beyond the age of 18 years was completely unaffected by birth cohort.13 For this reason, outcomes related to birth cohort were not the focus of this study. After obtaining institutional research ethics board approval, we used the database to identify adult patient cohorts with one of the following 6 diagnoses previously shown to contribute to a high burden of premature death in adulthood:5,6,8,9 (1) Eisenmenger syndrome; (2) unrepaired cyanotic heart defect; (3) complete transposition of the great arteries after atrial switch repair; (4) congenitally corrected transposition of the great arteries; (5) Fontan palliation for single ventricle physiology; and (6) repaired tetralogy of Fallot. For each cohort, we examined the evolution of numbers of patients and their age distribution between the years 1980 and 2009. Survival probabilities for each patient cohort were compared to the age- and gender-matched general Canadian population.14 Based on these survival probabilities, we predicted future numbers of deaths in the years 2010 through 2029.
Statistical Methods All analyses were performed in the R programming language.15 Patient characteristics have been presented as medians (with interquartile range) for continuous variables and as counts (with percentage) for nominal variables. Survival Analysis Survival of adult patients (≥18 years) within each of the 6 disease cohorts was estimated by computing Kaplan–Meier survival curves using age as time scale, with Canadian age- and gendermatched survival data used for comparison (18year-old Canadians in 1980 with the same sex distribution).14,16 Individual survival times were both left-truncated and right-censored, as is appropriate for such an observational study.16–18 In such Kaplan–Meier estimates, survival probability
Mortality Burden of Adults with Complex CHD is assumed to be independent of age at entry into the database; for example, a 35-year-old patient who was referred to the adult clinic at age 18 is assumed to have the same survival probability as a patient with the same diagnosis who was first referred at age 35. However, in clinical practice, patients who are referred to the adult clinic not through routine transition from pediatric care at around age 18 but rather later in life may be referred in response to a clinical complication, thus introducing bias. In order to minimize this bias, we computed survival estimates only for patients who entered the cohort at
ORIGINAL ARTICLES Increasing Mortality Burden among Adults with Complex Congenital Heart Disease Matthias Greutmann, MD,*†1 Daniel Tobler, MD,†‡1 Adrienne H. Kovacs, PhD,† Mehtap Greutmann-Yantiri, MD,† Sarah R. Haile, PhD,§¶ Leonhard Held, PhD,§ Joan Ivanov, MSc, PhD,†** William G. Williams, MD,†† Erwin N. Oechslin, MD,† Candice K. Silversides, MD,† and Jack M. Colman, MD† *Adult Congenital Heart Disease Program, University Hospital Zurich, Zurich, Switzerland; †Toronto Congenital Cardiac Centre for Adults, Peter Munk Cardiac Centre, Toronto General Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada; ‡Adult Congenital Heart Disease Clinic, University Hospital Basel, Basel, Switzerland; § Division of Biostatistics, Institute for Social and Preventive Medicine, University of Zurich, Zurich, Switzerland; ¶Clinical Trials Unit, Kantonsspital St. Gallen, St. Gallen, Switzerland; **Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada; ††Department of Cardiovascular Surgery, Hospital for Sick Children, Toronto, Ontario, Canada ABSTRACT
Background. Progress in management of congenital heart disease has shifted mortality largely to adulthood. However, adult survivors with complex congenital heart disease are not cured and remain at risk of premature death as young adults. Thus, our aim was to describe the evolution and mortality risk of adult patient cohorts with complex congenital heart disease. Methods. Among 12 644 adults with congenital heart disease followed at a single center from 1980 to 2009, 176 had Eisenmenger syndrome, 76 had unrepaired cyanotic defects, 221 had atrial switch operations for transposition of the great arteries, 158 had congenitally corrected transposition of the great arteries, 227 had Fontan palliation, and 789 had repaired tetralogy of Fallot. We depict the 30-year evolution of these 6 patient cohorts, analyze survival probabilities in adulthood, and predict future number of deaths through 2029. Results. Since 1980, there has been a steady increase in numbers of patients followed, except in cohorts with Eisenmenger syndrome and unrepaired cyanotic defects. Between 1980 and 2009, 308 patients in the study cohorts (19%) died. At the end of 2009, 85% of survivors were younger than 50 years. Survival estimates for all cohorts were markedly lower than for the general population, with important differences between cohorts. Over the upcoming two decades, we predict a substantial increase in numbers of deaths among young adults with subaortic right ventricles, Fontan palliation, and repaired tetralogy of Fallot. Conclusions. Anticipatory action is needed to prepare clinical services for increasing numbers of young adults at risk of dying from complex congenital heart disease. Key Words. Adult Congenital Heart Disease; Congenital Heart Surgery; Mortality; Outcomes
Introduction
C
omplex congenital heart defects affect about 0.6% of newborns, with stable incidence over time.1,2 Most affected patients are now expected to survive to adulthood.3,4 However, adults with repaired complex congenital heart lesions and with 1
These authors contributed equally to the study design, data interpretation, and manuscript preparation.
unrepaired cyanotic lesions (including large shunt lesions complicated by Eisenmenger syndrome) remain at risk for long-term complications and are at increased risk for death as young adults.5–8 Surgical repair techniques for complex congenital heart lesions were introduced several decades ago and have subsequently undergone continuous modification. Thus, cohorts of adult survivors are still emerging and evolving in terms
Congenit Heart Dis. 2015;10:117–127 © 2014 The Authors. Congenital Heart Disease Published by Wiley Periodicals, Inc. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
118 of numbers, age distribution, complexity, previous interventional management, and survival patterns. In previous analyses of deaths in adults with congenital heart disease, cohorts with Eisenmenger syndrome, unrepaired cyanotic heart defects, atrial switch operations for transposition of the great arteries, congenitally corrected transposition of the great arteries, Fontan palliation for univentricular physiology, and repaired tetralogy of Fallot contributed disproportionately to the number of young adults dying from their disease.5,6,8,9 Understanding the evolution of such adult patient cohorts and their mortality risk is essential for planning future care and research. The aims of this study were (1) to examine the 30-year evolution of 6 cohorts of adult survivors of common complex congenital heart lesions at a large tertiary care center for adults with congenital heart disease; (2) to analyze survival probabilities in these patient cohorts; and (3) to predict future trends in numbers of deaths over the upcoming two decades.
Methods
Database and Patient Cohorts The Toronto Congenital Cardiac Centre for Adults at Toronto General Hospital was founded in 1959. From its inception, it has collaborated closely with the Toronto Hospital for Sick Children (SickKids), which is the largest pediatric cardiology center in Canada. At SickKids, congenital cardiac surgery began in 1944, and the first open-heart operation using a heart–lung machine took place in 1958. SickKids surgeons have performed 80–90% of all congenital cardiac surgical procedures for the province of Ontario, Canada. Over the study period, the population in Ontario increased from 9 million to 13 million in 2009. The clinical and administrative database at the Toronto Congenital Cardiac Centre for Adults includes all adults with congenital heart disease ever followed by the program, whether strictly ambulatory or as inpatients or both. It contains information about the anatomical cardiac diagnoses and previous surgical and nonsurgical treatment history. As well, the database is used to track follow-up visits. If patients do not attend scheduled follow-up visits they and, as necessary, their alternate contacts and general practitioners are contacted by administrative staff. Survival status is regularly updated. The mode of death of patients Congenit Heart Dis. 2015;10:117–127
Greutmann et al. who died at our institution up to 1996 has previously been reported.6 In the past, survival of children with complex congenital heart disease was critically dependent on the availability of surgical repair at a given center. In the early days of surgical repair, improving perioperative care led to increased survival of affected children. In contrast, long-term survival of those who survived the operation and early complications, reaching adulthood, was not further impacted by era of birth cohort.10–12 Indeed, a recent study from Belgium demonstrated that survival beyond the age of 18 years was completely unaffected by birth cohort.13 For this reason, outcomes related to birth cohort were not the focus of this study. After obtaining institutional research ethics board approval, we used the database to identify adult patient cohorts with one of the following 6 diagnoses previously shown to contribute to a high burden of premature death in adulthood:5,6,8,9 (1) Eisenmenger syndrome; (2) unrepaired cyanotic heart defect; (3) complete transposition of the great arteries after atrial switch repair; (4) congenitally corrected transposition of the great arteries; (5) Fontan palliation for single ventricle physiology; and (6) repaired tetralogy of Fallot. For each cohort, we examined the evolution of numbers of patients and their age distribution between the years 1980 and 2009. Survival probabilities for each patient cohort were compared to the age- and gender-matched general Canadian population.14 Based on these survival probabilities, we predicted future numbers of deaths in the years 2010 through 2029.
Statistical Methods All analyses were performed in the R programming language.15 Patient characteristics have been presented as medians (with interquartile range) for continuous variables and as counts (with percentage) for nominal variables. Survival Analysis Survival of adult patients (≥18 years) within each of the 6 disease cohorts was estimated by computing Kaplan–Meier survival curves using age as time scale, with Canadian age- and gendermatched survival data used for comparison (18year-old Canadians in 1980 with the same sex distribution).14,16 Individual survival times were both left-truncated and right-censored, as is appropriate for such an observational study.16–18 In such Kaplan–Meier estimates, survival probability
Mortality Burden of Adults with Complex CHD is assumed to be independent of age at entry into the database; for example, a 35-year-old patient who was referred to the adult clinic at age 18 is assumed to have the same survival probability as a patient with the same diagnosis who was first referred at age 35. However, in clinical practice, patients who are referred to the adult clinic not through routine transition from pediatric care at around age 18 but rather later in life may be referred in response to a clinical complication, thus introducing bias. In order to minimize this bias, we computed survival estimates only for patients who entered the cohort at
