The Challenges of Managing Congenital Heart Disease After Childhood
With advances in the management of patients with congenital heart disease (CHD), the number of adults with CHD has grown substantially. In the decade after the first intracardiac repair of a congenital heart defect in 1952, congenital heart surgery was associated with short-term mortality rates of 60% to 70%.1 In the current era, mortality for even the most complex defects is 10% or less in high-volume congenital heart centers.2 As a result of both improved surgical care and advances in medical management, more than 80% of patients born with CHD can expect to live into adulthood.3 By 2020, it is projected that the majority of patients with CHD will be >18 years of age.4
Article p 2345
In this context, it is important to consider where and by whom these patients should be treated. Historically, CHD has been the domain of children’s hospitals with specialized resources for managing these conditions. At such centers, pediatric cardiologists, congenital heart surgeons, and anesthesiologists are arguably most familiar with the complex anatomy and physiology associated with CHD, as well as the most current treatment advances. Specialized intensive care units, nursing staffs, and radiology services are additional advantages of pediatric congenital heart centers. On the other hand, as patients with CHD move into adulthood, they acquire other medical conditions less familiar to physicians and staff at children’s hospitals, including obesity, hypertension, diabetes mellitus, and chronic lung disease. Effective management of such comorbidities may be equally important to enhancing patient longevity and quality of life and may be better entrusted to physicians and systems used to treating adults.
In this issue of Circulation, Karamlou et al5 help inform one crucial component of this broader debate: which types of surgeons should be operating on adults with CHD. The authors used data from the National Inpatient Sample to study operative mortality in adults undergoing an open heart surgery for 1 of 12 different types of CHD between 1988 and 2003. Overall, patients treated by pediatric heart surgeons (PHS) had substantially lower risk-adjusted mortality than those cared for by nonpediatric heart surgeons (NPHS; 1.9% versus 4.8%). In the study, PHS were defined as those surgeons for whom patients under 18 years of age constituted at least 75% of their operative practice. At the hospital level, centers with higher pediatric case volumes also had better survival rates. On the basis of these findings, the authors conclude that the surgical care of adults with CHD should be concentrated in the hands of PHS.
Although the findings of this study are both novel and compelling, differences in operative mortality rates for PHS and NPHS may be confounded in part by differences in patient case mix and other factors. The limitations of administrative data in capturing comorbidities and illness severity for purposes of risk adjustment are well known. Administrative data would not accurately reflect preoperative ventricular function, the revisionary nature of surgery, or the underlying indication for surgery. For example, a young adult with a restrictive perimembranous ventricular septal defect and a 75-year-old with an infarction-induced ventricular septal defect have very different risk profiles. Moreover, PHS and NPHS may tend to do different types of procedures associated with different baseline risks. In the present study, the large majority (80%) of cases performed by NPHS involved ostium secundum defects, a relatively simple operation. PHS performed fewer such operations but were more likely to perform procedures for tetralogy of Fallot and other more complex procedures. Although the authors attempted to adjust for underlying diagnoses, comparisons of procedure-specific mortality would have been informative.
In interpreting specialty-related differences in outcomes, it is essential to consider potential differences in the systems in which different types of surgeons work. The authors assert the most obvious and intuitive explanation for their findings: that PHS have more experience in CHD surgery, which leads directly to greater expertise in decision making and technical proficiency and, ultimately, better outcomes. There is ample support in the extensive volume-outcome literature for this “practice-makes-perfect” hypothesis. However, there is also strong evidence that hospital-level characteristics, including volume, may be even stronger determinants of operative outcomes than surgeon volume alone for some procedures, including CHD.6 In this context of the study herein, it is likely that PHS and NPHS work in different settings with different priorities, resources, and expertise. To the extent that they work in more specialized centers, PHS may have superior outcomes in part because they get better support from multidisciplinary teams, specialized cardiac intensive care units, pediatric cardiac anesthesiologists, and other colleagues. In future work, the authors could extend their analysis using hierarchical models to disentangle the contributions of hospital- and surgeon-level factors to patient outcomes.
Although it remains unclear whether such strategies would be best based on hospital or surgeon criteria, findings from the study by Karamlou et al5 suggest the potential value of concentrating the care of adults with CHD among selected providers. Interest in selective referral strategies for surgical care is growing. The Leapfrog Group (www.leapfroggroup.org), a large coalition of public and private purchasers, has targeted 7 complex procedures for “evidence-based hospital referral,” guided by volume standards and other criteria. Other large private payers are implementing similar Centers of Excellence programs for adult cardiac surgery and other types of specialty care. Although the volume-outcome literature supports the potential benefits of such strategies for patients, the value of selective referral remains uncertain. First, although volume, specialty type, and other structural variables are often useful in identifying groups of providers with above-average outcomes, they perform poorly in predicting the performance of individual hospitals or surgeons. Second, there would be many practical challenges associated with selective referral strategies for adults with CHD. With just over 100 congenital cardiac centers nationwide, it is not clear whether such facilities have sufficient capacity to accommodate adult referrals from the 1000 hospitals with general cardiac programs. Concentrating care within specialized centers may also imply unacceptable travel burdens and access problems in many areas of the country.7 Finally, if efforts aimed at concentrating care were limited to the surgical episode, selective referral might imply more fragmented care for adults with CHD.
Given the many challenges inherent with selective referral strategies, efforts aimed at improving quality in all settings are essential. Ensuring that cardiothoracic surgeons are adequately trained in congenital heart surgery is one obvious approach to achieving this goal. The American Board of Thoracic Surgeons recently approved board-eligible residency programs in congenital heart surgery, but there are no specific requirements for adult CHD procedures. To better meet the needs of the growing number of adults with CHD, the American Board of Thoracic Surgeons should develop more explicit standards for ensuring an adequate operative experience and knowledge base for trainees in general cardiac programs. Achieving broad-based quality improvement will also require a more robust national platform for tracking outcomes with CHD surgery. The national cardiac surgery registry of the Society of Thoracic Surgery has achieved nearly full participation among US hospitals involved in CABG and heart valve replacement. The Society of Thoracic Surgery has recently added a new registry for congenital heart surgery. As hospital participation grows, the congenital heart registry may become invaluable in supporting local quality improvement. In addition, its outcome measures may ultimately replace structural measures, including surgeon specialty, for profiling provider quality.
Although better training and better outcomes data are essential, improving the quality of care for adults with CHD will require research aimed at elucidating the processes of care that underlie variation in outcomes. In other words, we need a clearer sense of what PHS and their institutions are doing differently to achieve lower mortality rates. High-leverage processes might include those associated with preoperative decision making and the timing of surgical intervention, eg, when to proceed with pulmonary valve replacement for patients with tetralogy of Fallot. A better understanding of the relative effectiveness of different procedures and techniques would also be invaluable. Finally, understanding best practices related to medical management would be essential for optimizing patients’ clinical status before surgery and avoiding adverse events afterward. The study by Karamlou et al5 highlights the problem of variation in provider outcomes with CHD surgery and thus potential opportunities for improvement. They and other clinical researchers should take the next step in identifying mechanisms underlying variation and making improvement a reality.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
Macmahon B, McKeown T, Record RG. The incidence and life expectation of children with congenital heart disease. Br Heart J. 1953; 15: 121–129.
British Cardiac Society. Grown-up congenital heart (GUCH) disease: current needs and provision of service for adolescents and adults with congenital heart disease in the UK: report of the British Cardiac Society Working Party. Heart. 2002; 88 (suppl 1): i1–i14.
Karamlou T, Diggs BS, Person T, Ungerleider RM, Welke KF. National practice patterns for management of adult congenital heart disease: operation by pediatric heart surgeons decreases in-hospital death. Circulation. 2008; 118: 2345–2352.
Dimick JB, Finlayson SR, Birkmeyer JD. Regional availability of high-volume hospitals for major surgery. Health Aff (Millwood). 2004; (suppl Web exclusives): VAR45–VAR53.