Adults or Big Kids: What Is the Ideal Clinical Environment for Management of Grown-Up Patients With Congenital Heart Disease?

Transcription

1 ORIGINAL ARTICLES: SURGERY: The Annals of Thoracic Surgery CME Program is located online at To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal. Adults or Big Kids: What Is the Ideal Clinical Environment for Management of Grown-Up Patients With Congenital Heart Disease? Tara Karamlou, MD, Brian S. Diggs, PhD, Ross M. Ungerleider, MD, MBA, and Karl F. Welke, MD, MS Divisions of Cardiothoracic Surgery and Surgery, Oregon Health and Science University, Portland, Oregon; Division of Pediatric Cardiothoracic Surgery, Department of Surgery, Case Western Reserve University, Cleveland, Ohio; and Mary Bridge/Swedish Pediatric Cardiothoracic Surgery Program, Mary Bridge Children s Hospital and Health Center, Multicare Health System, Tacoma, Washington Background. Initiatives to develop Adult Congenital Centers for management of grown-up congenital heart disease (GUCH) patients (aged >18 years) have widened without evidence identifying the ideal clinical environment. To elucidate the optimum care paradigm, we investigated whether mortality for patients with GUCH was influenced by the type of hospital where they had surgery, children s specialty hospital (CH) versus general hospital (GH), and by the clinical focus of the surgeon, congenital heart surgery (CHS) or noncongenital (adult acquired) heart surgery (NCHS). Methods. In the Nationwide Inpatient Sample , we identified index procedures in patients 18 or more years of age within 12 congenital cardiac disease diagnostic groups. The CHS surgeons were defined as those whose annual practice volume consisted of more than 75% pediatric cardiac operations. Four clinical environment combinations were constructed: CH plus CHS, CH plus NCHS, GH plus CHS, and GH plus NCHS. Years were grouped into quartiles to identify trends in management over time. Results. In all, 29,070 operations occurred at GH and 10,971 occurred at CH. Unadjusted in-hospital mortality was lowest in the CH plus CHS environment (1.14%), and highest for in the GH plus CHS environment (9.93%; p < 0.001). After risk adjustment for patient factors, the CH plus CHS environment remained optimum, whereas the other three environments increased the risk of inhospital death (GH plus NCHS: odds ratio 2.4 [95% confidence interval: 0.9 to 6.2]; CH plus NCHS: odds ratio 2.4 [95% confidence interval: 0.9 to 6.5]; GH plus CHS: odds ratio 9.1 [95% confidence interval: 3.0 to 27.6]). Over the study period, there was a dramatic rise in the number of GUCH patients treated in GH plus NCHS and CH plus NCHS, suggesting that the shift in clinical environment was provider specific rather than hospital-type specific. Conclusions. Case mix varies with the clinical environment, with more complex procedures performed at GH plus CHS. The optimal environment for complex GUCH surgery involved CHS operating within CH. Initiatives to develop adult congenital centers dedicated to the care of GUCH patients are warranted, and should include congenital heart surgeons operating in a setting mimicking children s hospitals. (Ann Thorac Surg 2010;90:573 9) 2010 by The Society of Thoracic Surgeons Owing to the improved outcomes for congenital heart surgery in the neonatal period, there is an increasing number of patients who are grown-ups with congenital heart disease (GUCH) [1 5]. The best estimates of the current population of GUCH patients came from the Accepted for publication Feb 26, Presented at the Forty-fifth Annual Meeting of The Society of Thoracic Surgeons, San Francisco, CA, Jan 26 28, Address correspondence to Dr Karamlou, Oregon Health and Science University, Division of Cardiothoracic Surgery, 3181 SW Sam Jackson Park Rd, Mail Code L-353, Portland, OR 97239; karamlou@ ohsu.edu. 32nd Bethesda Conference held in 2000, which projected that nearly 787,800 adult patients were living with congenital heart disease [1]. Many GUCH patients require ongoing specialized care and surgical intervention, and despite the notion of adulthood, GUCH patients still offer unique physiologic and anatomic challenges consistent with congenital rather than adult acquired heart disease. However, as adults, GUCH patients may have important comorbidities such as renal failure, peripheral vascular disease, chronic obstructive pulmonary disease, deep venous thrombosis, and substance abuse, which are uncommon among children [6 8]. Rossano and col by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 574 KARAMLOU ET AL Ann Thorac Surg GROWN-UPS WITH CONGENITAL HEART DISEASE 2010;90:573 9 Abbreviations and Acronyms ASD atrial septal defect CH children s hospital CHS congenital heart surgeon GH general hospital GUCH grown-up congenital heart disease NCHS noncongenital heart surgeon NIS Nationwide Inpatient Sample VSD ventricular septal defect leagues [8] have shown that postoperative adverse events occur more frequently when GUCH patients are operated on at children s hospitals, especially in the setting of comorbidities. Because of these unique challenges that characterize GUCH, there has been a growing focus on developing specialized centers dedicated to the care of this population [1 4, 9 13]. Many large, academic centers have established so-called adult congenital centers that offer comprehensive care by cardiologists and surgeons who are familiar with the pathology, anatomy, and social issues facing GUCH patients [1 4, 12, 13]. We showed [14] that specialist care for GUCH patients reduced in-hospital mortality without a corresponding increase in hospital length of stay and total hospital charges. However, incompletely evaluated in our prior study was the influence of the hospital environment on both mortality and resource utilization. That is, congenital heart disease surgeons and adult acquired heart disease surgeons generally operate in different environments, and the surgeon themselves may in fact be surrogates for the systems or environment that they bring with them. We therefore undertook this study to determine national practice patterns for clinical environments utilized for the care of GUCH patients and to determine the ideal clinical environment for GUCH patients that would reduce inhospital mortality. Material and Methods Patient and Surgeons The Nationwide Inpatient Sample (NIS) is a stratified, cross-sectional database that includes approximately 20% of all community (nonfederal) hospital discharges in the United States. The NIS is managed under the Healthcare Cost and Utilization Project and is stratified by geographical region, hospital bed size, teaching status, urban versus rural location, and hospital ownership. This study s data were derived from the combination of the NIS databases from 1988 to 2003, and our analysis used sampling weights provided within the NIS to derive national estimates. The sampling frame of the NIS has been nearly constant over time except for a modification in 1998, which excluded short-term rehabilitation hospitals from the sampling frame. To account for this small change, we used the revised NIS trend weights, published by the Agency for Healthcare Research and Quality [15], which allows one to compare across years with the same effective sampling scheme. The NIS was searched for the individual years 1988 through 2003, selecting out hospital discharges for which the primary International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) procedure codes were for all open-heart or thoracic aortic (eg, coarctation of the aorta) operations [14]. (A complete list of codes utilized can be found at: codes_ :_congenital_anomalies#circulatory_ system.) Comorbidities, based on the methodology described by Elixhauser and associates [16], were also abstracted. Hospital discharges were excluded if the surgeon identifier was missing. The Institutional Review Board at Oregon Health and Sciences University provided approval for this study and expedited review owing to the deidentified nature of the dataset. Using this initial procedural dataset, we then incorporated 12 ICD-9-CM coded diagnoses adjudicated by consensus among the authors that capture the vast majority of GUCH [14]. However, because the Agency for Healthcare Research and Quality requirements prohibit reporting of cases with less than 10 counts, the following diagnoses were combined into a larger category labeled all other diagnoses : hypoplastic left heart syndrome, D-transposition of the great arteries, L-transposition of the great arteries, double-outlet right ventricle, total anomalous pulmonary venous return, and ostium primum atrial septal defect. The GUCH operations were then defined as those operations within these specified 12 diagnoses performed in patients aged 18 years or more. We then profiled individual surgeons by previously described methods [15]: surgeons were identified within the NIS based on the synthetic surgeon identifier (surgeon_id), which is distinct from the attending physician identifier (phy_id). Individual surgeons are unable to be traced over time within the NIS, and therefore we expressed concentration in pediatric heart surgery as a percentage of any individual surgeon s annual cardiac caseload. Congenital heart surgeons (CHS) were defined as surgeons whose annual practice consisted of 75% or more pediatric heart surgery (eg, operations undertaken in patients aged less than 18 years). Noncongenital heart surgeons (NCHS) were primarily adult acquired heart disease surgeons, therefore, whose annual practice volume consisted of less than 75% pediatric heart surgery. The 75% cutpoint was chosen based upon our prior publication elucidating the relationship between outcomes and surgeon specialty [15]. Owing to the nature of the NIS, whereby different hospitals are sampled in any given year, surgeon classification was based on annual volume for each individual year (ie, year specific). Practice setting was defined based on the National Association of Children s Hospitals and related institutions (NACH) designation given to the hospital. This designation does not appear directly in the NIS; it was extracted from contemporaneous years of the Kids Inpatient Database [17]. The National Association of Children s Hospitals designates hospitals as not identified as a children s hospital, children s general hospital,

3 Ann Thorac Surg KARAMLOU ET AL 2010;90:573 9 GROWN-UPS WITH CONGENITAL HEART DISEASE 575 Table 1. Comparison of Patients and Hospital Characteristics Among Clinical Environments Variable CH CHS CH NCHS GH CHS GH NCHS Number of patients 327 1, ,672 Demographics Mean age SEM, years Female, % SD Diagnosis, weighted frequency SD (%) PAVSD (1.5) Not reportable Not reportable (0.2) Tetralogy of Fallot (20) (2.8) (22) (1.4) Coarctation of the aorta (6.0) (4.2) (7.1) 1, (3.8) VSD (26.1) 1, (13.7) (15.0) 3, (13.3) Ostium secundum ASD (46.6) 7, (77.4) (41.1) 23, (80.4) All other diagnoses a (22.4) (5.1) (18.4) (3.4) Comorbidities Congestive heart failure Not reportable (1.2) Not reportable (0.9) PVD Not reportable (2.8) Not reportable 1, (3.7) Renal failure 10 7 (0.6) (1.3) Not reportable (1.6) Complicated DM Not reportable (0.9) Not reportable (1.1) Chronic lung disease (3.0) (8.3) 12 8 (3.6) 2, (9.4) Obesity 21 9 (1.3) (2.6) Not reportable (3.1) a Other diagnoses includes the following: hypoplastic left heart syndrome, total anomalous pulmonary venous return, ostium primum atrial septal defect, D-transposition of the great arteries, L-transposition of the great arteries, and double-outlet right ventricle. ASD atrial septal defect; CH children s hospital; CHS congenital heart surgery; DM diabetes mellitus; GH general hospital; NCHS noncongenital heart surgery; PAVSD pulmonary atresia with ventricular septal defect; PVD peripheral vascular disease; VSD ventricular septal defect. children s specialty hospital, and children s unit in a general hospital. No children s specialty hospitals were present in these data. For the purposes of this analysis, children s general hospital and children s unit in a general hospital were considered equivalent (CH), and those not identified as a CH were considered general hospitals (GH). Cases in which we could not identify the practice setting (n 74) were excluded from the analysis. The surgeon classification and practice setting were combined to give four clinical environments: CH plus CHS, GH plus CHS, CH plus NCHS, and GH plus NCHS. Statistical Analysis The primary outcome was in-hospital mortality. Logistic regression models, adjusted for patient and hospital level characteristics, were fit to estimate the attributable risk as a result of operation within each of the four environments, and within children s hospitals compared with general hospitals (eg, irrespective of the surgeon specialty). We generated c-indices and receiver-operating characteristic curves for all logistic regression models to provide insight into model discrimination. We used SAS software, version 9.1 (SAS Institute, Cary, NC) to fit the models, accounting for the survey design of the NIS, the potential clustering of outcomes within a hospital, and the decrease in clustering occurring over increasing time intervals. Years were grouped into quartiles to identify trends over time. Our method of risk adjustment included fitting an initially saturated logistic regression model, using the following variables: clinical environment, era (year quartiles), all comorbidity measures, female sex, patient age (as a continuous variable and as a ranked quartile variable), and patient primary diagnostic group (tetralogy of Fallot, atrial septal defect [ASD], ventricular septal defect [VSD], or other). Manual stepwise backward selection was used excluding those variables with nonsignificant p values. Minimization of the information criteria (AIC, BIC) were used to evaluate model fit after exclusion of each variable, and the full model c-statistic was also assessed. Interaction terms Fig 1. Case mix among the four clinical environments varied significantly. In general, the congenital heart surgeons (CHS) operating within either practice setting (children s hospitals (CH) or general hospitals (GH)) performed fewer atrial septal defect (ASD) repairs and more complex operations. On the other end of the spectrum, the noncongenital heart surgeons (NCHS) operating within either practice setting performed more than 80% ASD repairs with few complex procedures. (Blue sections ASD; green sections tetralogy of Fallot; red sections ventricular septal defect; purple sections other.)

4 576 KARAMLOU ET AL Ann Thorac Surg GROWN-UPS WITH CONGENITAL HEART DISEASE 2010;90:573 9 Table 2. Multivariable Factors Associated With In-Hospital Mortality for Model With Four Clinical Environments Variable Clinical environment CH CHS Odds Ratio 95% CI p Value Reference category CH NCHS GH NCHS GH CHS Age (per year) Diagnosis Ostium secundum ASD Reference category Tetralogy of Fallot VSD Pulmonary atresia/vsd Coarctation of the aorta All other diagnoses Comorbidities Congestive heart failure Renal failure Complicated diabetes mellitus Era Quartile 1 Reference category Quartile Quartile Quartile ASD atrial septal defect; CH children s hospital; CHS congenital heart surgery; CI confidence interval; GH general hospital; NCHS noncongenital heart surgery; VSD ventricular septal defect. prevalence of congestive heart failure, renal failure, diabetes mellitus, and obesity were statistically similar among all groups (Table 1). Case mix varied significantly among the four clinical environments (Fig 1), with the GH plus CHS group performing fewer ASD closures and more complex operations (p 0.001). Importantly, the overwhelming majority of cases (80%) performed by adult acquired surgeons were ASD closures, irrespective of hospital type. Note, however, that the designation transposition of the great arteries, tetralogy of Fallot, or double-outlet right ventricle represent the patient s primary congenital diagnosis, but as expected, the vast majority of these adult patients underwent post-repair procedures (eg, pulmonary valve replacements, ablation of arrhythmias, VSD closure) rather than primary repair. Trends Over Time There was a significant increase in the number of GUCH cases over time, with 564 of the identified cases (national estimate 3, [8%]) performed in the earliest quartile (1988 to 1991) and 3,251 cases (national estimate 15, [40%]) performed in the most recent quartile (2000 to 2003). Examined by hospital type, we found no significant increase in the proportion of patients treated at general hospitals compared with children s hospitals over the duration of the study period (Fig 2). However, we did find a significant change in treatment of GUCH patients when we examined the four clinical environments. Specifically, we found a dramatic rise in the number of patients treated in GH plus NCHS and CH plus NCHS, suggesting that the shift in clinical environment is provider specific rather than hospital-type specific (Fig 3). were considered, and transformations of scale were used to improve calibration. The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agreed to the manuscript as written. Results We identified 7,863 operations between 1988 and 2003, yielding a national estimate of 40,041 1,339 operations. Of these, 327 (national estimate 1, [4%]) were CH plus CHS, 1,800 (national estimate 9, [23%]) were CH plus NCHS, 59 (national estimate [1%]) were GH plus CHS, and 5,672 (national estimate 28, [72%]) were GH plus NCHS. Demographic data regarding patients and hospitals are shown in Table 1. Consistent with our prior report, we found that patients operated on by NCHS were older (mean age 48 years) compared with patients operated on by CHS (mean age 28 years, p 0.001), irrespective of hospital type. Prevalence of peripheral vascular disease and chronic lung disease was highest among patients operated on at GH plus NCHS (3.7% [p 0.01] and 9.4% [p 0.001], respectively) compared with the other three groups, but Fig 2. A significant increase in the number of grown-up congenital heart disease (GUCH) cases occurred over time, with 564 cases (3, [8%]) performed in the earliest quartile (1988 to 1991) and 3,251 cases (15,895 1,010 [40%]) performed in the most recent quartile (2000 to 2003). Examined by hospital type, however, we found no significant increase in the proportion of patients treated at general hospitals (gray bars) compared with children s hospitals (black bars) over the duration of the study period.

5 Ann Thorac Surg KARAMLOU ET AL 2010;90:573 9 GROWN-UPS WITH CONGENITAL HEART DISEASE 577 lowest risk of in-hospital death (Table 2). Risk of inhospital death was nearly equivalent among CH plus NCHS and GH plus NCHS (odds ratios 2.43 and 2.35, respectively), but remained lower than for GH plus CHS (p 0.001). Other factors increasing the risk of in-hospital death included older age, presence of congestive heart failure, diabetes mellitus, or renal failure, and more complex diagnostic groups such as tetralogy of Fallot, pulmonary atresia with VSD, and VSD. Operation within the recent era was protective. Model discrimination was excellent, with a c-statistic of Fig 3. There was a significant change in treatment of grown-up congenital heart disease (GUCH) patients when we examined the four clinical environments. Specifically, we found a dramatic rise in the number of patients treated by noncongenital heart surgeons (NCHS), irrespective of hospital type either children s hospital (CH) or general hospital (GH) suggesting that the shift in practice setting is provider specific rather than hospital-type specific. In contrast, the number of GUCH patients operated on by congenital heart surgeons (CHS) in either practice setting remained fairly constant. (Blue bars CH CHS; gray bars CH NCHS; open bars GH CHS; black bars GH NCHS.) Comment We have characterized the national practice patterns over time for GUCH patients among four clinical environments in the United States. Our study found that GUCH patients fare best when treated by congenital heart surgeons in a children s hospital. In-hospital mortality was reduced fourfold within the CH plus CHS environment compared with the next best environment, CH plus NCHS. We hypothesized, based upon our earlier report [15], that decreased in-hospital mortality among GUCH patients was likely the result of refined processes of care at institutions possessing specialized congenital heart surgeons (eg, the children s hospital environment) in addition to the expertise of the surgeons themselves. Data from the present study would seem to confirm this hypothesis, although we acknowledge that important differences in patient age and case-mix existed among provider types, and we cannot completely adjust for these within the context of a retrospective study. For instance, we believe that increased mortality among the GH plus CHS environment may be secondary to the increased complexity of patients within this subgroup. It In-Hospital Mortality National estimated overall in-hospital mortality for GUCH patients was 1,874 (95% confidence interval: 1,645 to 2,104 [4.7% (4.2% to 5.2%)]). Estimated in-hospital mortality for GUCH patients operated on in CH plus CHS (1.1% [0.1% to 2.1%]) was lower than all other clinical environments, whereas estimated in-hospital mortality was highest for GUCH patients operated on in GH plus CHS (9.9% [5.3% to 5.7%]; p 0.001; Fig 4). Examined by hospital type, there was a trend toward improved survival for GUCH patients operated on in CH compared with patients operated on in GH (3.9% [2.9% to 4.8%] versus 5.0% [4.3% to 5.8%]; p 0.06). Accordingly, GUCH patients within CH plus NCHS had a small reduction in mortality (4.3% [3.3% to 5.4%]) compared with GUCH patients in GH plus NCHS (5.0 [4.2% to 5.7%]; p 0.35), suggesting a possible benefit to the children s hospital environment. A multivariable logistic regression model adjusted for era, patient, and hospital characteristics showed that GUCH patients operated on in CH plus CHS had the Fig 4. Estimated in-hospital mortality for grown-up congenital heart disease (GUCH) patients operated on within children s hospital (CH) and by congenital heart surgeons (CHS) was lower (1.1% [0.1% to 2.1%]) than for all other clinical environments, whereas estimated in-hospital mortality was highest for GUCH patients operated on within GH plus CHS (9.9% [5.3% to 5.7%]; p 0.001). Examined by hospital type, there was a trend toward improved survival for GUCH patients operated on in CH compared with patients operated on in GH (3.9% [2.9% to 4.8%] versus 5.0% [4.3% to 5.8%]; p 0.06). Accordingly, GUCH patients in CH plus NCHS had a small reduction in mortality (4.3% [3.3% to 5.4%]) compared with GUCH patients in GH plus NCHS (5.0 [4.2% to 5.7%]; p 0.35), suggesting a possible benefit to the children s hospital environment.

6 578 KARAMLOU ET AL Ann Thorac Surg GROWN-UPS WITH CONGENITAL HEART DISEASE 2010;90:573 9 is likely that for simpler lesions, such as secundum ASD, the provision of a specialized care setting may be less important. Our analysis supports this contention as mortality risk in GUCH patients with a diagnosis of ASD was not dependent on the care environment. We have also shown an increase in the number of GUCH patients operated on by NCHS over time irrespective of hospital type, whereas the prevalence of GUCH patients operated by CHS remained fairly constant. This trend is confirmed by our analysis of hospital type, which showed no significant increase in the proportion of patients treated at general hospitals compared with children s hospitals over the duration of the study period, suggesting that the shift in clinical environment is provider specific rather than hospital-type specific. This rise in treatment within an environment excluding CHS is particularly concerning in light of our prior report showing that operation by NCHS increases both inhospital mortality and resource utilization [15]. The subject of the ideal clinical environment in which to care for GUCH patients has been incompletely evaluated, despite the rapidly growing GUCH population [6 8, 18]. Rossano and colleagues [8] recently described the risk of adverse events among 135 GUCH patients undergoing cardiac surgery at Texas Children s Hospital from 2000 to They found acceptable early mortality (1.5%), but a 24% prevalence of adverse events. We cannot directly compare the results of this study within the context of our present report, however, as management of GUCH patients at Texas Children s Hospital has evolved to include a multidisciplinary team of congenital heart surgeons and both adult and pediatric medical specialists [7]. A care paradigm involving a combination of specialists from both the adult and the pediatric realm is one possible solution, but may present unreasonable hurdles in institutions without a nearby or even adjoining children s hospital. Mahle and colleagues [6] used the Pediatric Health Information System database to determine outcomes among 719 adult patients (aged 21 years) treated at 37 children s hospitals in the United States. Overall mortality for their study population was 1.6%, although nearly half of the procedures identified in this study (n 352) were pacemaker implantations/ revisions or other. Importantly, the prevalence of GUCH cases at the institutions sampled was uniformly low, ranging from 0% to 10.9%. These estimates agree with our findings that the minority of GUCH patients (less than 30%) even in the present era are cared for within a children s hospital setting. Given the data from the present study along with those from our prior report [15], such a trend is concerning. Limitations Our study has several limitations, many of which have been extensively covered in our prior reports [15, 19]. Concatenation of the current dataset required combining data from two large administrative datasets, which may inflate the prevalence of coding errors. Additionally, we could not include data from 74 admissions because of an inability to determine hospital type, raising the possibility that selection bias may have influenced our findings. Furthermore, because of the lack of time-date stamping, it is not possible to determine whether comorbidities identified in the NIS are admission diagnoses or after admission (or postprocedure) complications. Although it is not inherently problematic to compare populations of different sizes, large reduction in sample size, especially among the GH plus CHS group, which consisted of only 59 admissions, may have limited the power of our study by reducing the variability within this group. In conclusion, despite these limitations, our study has demonstrated that in-hospital mortality for GUCH patients is lowest in a clinical environment where congenital heart surgeons operate within a children s hospital. Survival for GUCH patients is, therefore, influenced by both the specialization of the surgeon and the care setting. Dramatic increases in the number of GUCH patients cared for by surgeons without specialization in congenital heart surgery are not ideal and may contribute to worsening outcomes among this challenging population. Initiatives to develop adult congenital centers dedicated to the care of complex GUCH patients are warranted, and should include congenital heart surgeons operating in a setting mimicking children s hospitals. References 1. Warnes CA, Liberthson R, Danielson GK, et al. Proceedings of the 32nd Bethesda conference: care of the adult with congenital heart disease. J Am Coll Cardiol 2001;37: Somerville J. Grown-up congenital heart disease medical demands look back, look forward Thorac Cardiovasc Surg 2001;49: Niwa K, Perloff JG, Webb GD, et al. Survey of specialized care facilities for adults with congenital heart disease. Int J Cardiol 2004;96: Webb GD. Care of adults with congenital heart disease a challenge for the new millennium. Thorac Cardiovasc Surg 2001;49: Reid GJ, Irvine MJ, McCrindle BW, et al. Prevalence and correlates of successful transfer from pediatric to adult health care among a cohort of young adults with complex congenital heart defects. Pediatrics 2004;113:e Mahle WT, Kirshbom PM, Kanter KR, Kogon BM. Cardiac surgery in adults performed at children s hospitals: trends and outcomes. J Thorac Cardiovasc Surg 2008;136: Mott AR, Fraser CD, McKenzie ED, Bezold LI, et al. Perioperative care of the adult with congenital heart disease in a free-standing tertiary pediatric facility. Pediatr Cardiol 2002; 23: Rossano JW, Smith EOB, Fraser CD, et al. Adults undergoing cardiac surgery at a children s hospital: an analysis of perioperative morbidity. Ann Thorac Surg 2007;83: Landzberg MJ, Murphy DJ, Davidson WR, et al. Task Force 4: organization of delivery systems for adults with congenital heart disease. J Am Coll Cardiol 2001;37: Therrien J, Dore A, Gersony W, et al. Canadian Cardiovascular Society consensus conference 2001 update. Recommendations for the management of adults with congenital heart disease. Part I. Can J Cardiol 2001;17: Therrien J, Gatzoulis M, Graham T, et al. Canadian Cardiovascular Society consensus conference 2001 update. Recom-

7 Ann Thorac Surg KARAMLOU ET AL 2010;90:573 9 GROWN-UPS WITH CONGENITAL HEART DISEASE 579 mendations for the management of adults with congenital heart disease. Part II. Can J Cardiol 2001;17: Therrien J, Warns C, Daliento L, et al. Canadian Cardiovascular Society consensus conference 2001 update. Recommendations for the management of adults with congenital heart disease. Part III. Can J Cardiol 2001;17: Gatzoulis MA, Hechter S, Stu SC, Webb GD. Outpatient clinics for adults with congenital heart disease: increasing workload and evolving patterns in referral. Heart 1999;81: Karamlou T, Diggs BS, Person T, Ungerleider RM, Welke KF. National practice patterns for management of adult congenital heart disease: operations by pediatric heart surgeons decreases in-hospital mortality. Circulation 2008;118: HCUP Nationwide Inpatient Sample (NIS). Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality, Rockville, MD. Available DISCUSSION DR RICHARD N. GATES (Orange, CA): I think the conclusions reached here are something that would play well with this audience, that is, congenital heart surgeons performing operations in a congenital environment are probably doing a better operation than adult heart surgeons doing congenital operations in an adult environment. But at least in our experience, patients in their 30s, 40s, and 50s want to be operated on in an adult environment. These patients are much higher risk as they frequently have poor ventricles and have been repaired in a different era than the patients who currently are 16 to 20 years old who are appropriately operated on in a children s hospital. So I m a little concerned about your conclusions because we do not have a STS congenital database predictive for adult congenital heart work. Since none of your patients are risk stratified, I believe that you ve shown us that low-risk congenital cases in young patients can be done with low risk in children s hospitals and that high-risk older patients can be done in adult hospitals with a high risk. So I d like you to address that. Secondly, was there any consideration to having a category for surgeons that are bona fide adult congenital surgeons, meaning that they performed, say, more than 150 adult and more than 150 congenital procedures per year? High-volume adult congenital surgeons don t fit in your category of congenital heart surgeons ( 75%) because, ironically, they do too much adult congenital surgery. These few bona fide adult congenital surgeons are lumped in with literally thousands of adult surgeons doing a few atrial septal defects a year. Did your analysis allow you to look at bona fide congenital heart surgeons outcomes compared with the other two categories of surgeons? Finally, what about the concept of a true adult congenital environment that s a combination facility, where all adult s and at: Accessed July 1, Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care 1998;36: HCUP Kids Inpatient Database (KID). Healthcare Cost and Utilization Project (HCUP), 2000 and Agency for Healthcare Research and Quality, Rockville, MD. Available at: Accessed July 1, Dearani JA, Connolly HM, Martinez R, Fontanet H, Webb GD. Caring for adults with congenital cardiac disease: successes and challenges for 2007 and beyond. Cardiol Young 2007;17(Suppl 2): Welke KF, Karamlou T, Diggs BS. Databases for assessing the outcomes of the treatment of patients with congenital and paediatric cardiac disease a comparison of administrative and clinical data. Cardiol Young 2008;18: all children s operations are performed in the same operating room by a common team of anesthesiologists, surgeons, scrubs, and perfusionists. There are a few dozen such institutions in the United States and many are members of the American Congenital Heart Association. Did you consider comparing ACHA program results with your two hospital group s results? DR KARAMLOU: Well, I ll take your last question first since it s foremost in my mind. We cannot answer that question with this data set. We have no idea which perfusionist did what and which anesthesiologist did what in these patients. Although I think your point is well taken, and this study by no means says that congenital heart surgeons in a children s hospital have to be the paradigm. I m simply stating that in lieu of developing dedicated adult congenital centers, of which there are very few, especially in the United States, that is a model that based on this data set yields the best results. And in regard to your other question, looking at dedicated adult congenital surgeons, that is something that we did in our original paper that was published in Circulation earlier this last year, and it s something that we could certainly extend to this hospital-level analysis. The problem with that is defining the surgeons who do enough adult congenital cases within this data set, and that proved problematic in our earlier analysis, although it s something that we could certainly look at. Regarding your first question, I think it s very difficult. The STS congenital database right now does not include codes for adult congenital heart disease cases, and I think that we ought to develop initiatives to look at specifically this growing population of patients. And something like you re suggesting, I think, would take that type of data set to answer.