Assessment of a congenital heart surgery programme: a reappraisal

Assessment of a congenital heart surgery programme: a reappraisal Abstract OBJECTIVES To assess our practice according to the Society of Thoracic Surgeons and the European Association for Cardio-Thoracic Surgery (STS-EACTS) Mortality Score and to the new concept of unit performance. METHODS All main procedures carried out in the years 2012–2016 were analysed. The STS-EACTS model-based mortality risk procedure was used to calculate expected mortality. Surgical performance was estimated as the Aristotle complexity score multiplied by hospital survival. Unit performance was defined as surgical performance multiplied by the number of procedures. RESULTS In total, 2435 procedures were analysed. One hundred and two deaths (95% confidence interval 71–135 deaths) were expected; 43 patients died after operation. Observed mortality divided by expected mortality was 0.42. The ratio ranged from 0.23 (year 2014) to 0.59 (year 2013) and was <0.6 in all STS-EACTS mortality categories. The difference between observed and expected mortality was highly significant: 1.8% vs 4.2% (P-value <0.0001). Observed surgical and unit performances were, higher than expected performances every year. Achieved surgical performance was the highest in year 2012 (7.28 ± 2.54) and the lowest in year 2014 (7.04 ± 2.52). The highest figure of unit performance was achieved in year 2016: 3980 points. CONCLUSIONS The STS-EACTS score, currently recognized as a sound instrument to assess mortality after congenital heart surgery, overestimated postoperative mortality. If these results are confirmed by other centres, the model should be recalibrated to match the current surgical practice. Although surgical performance can evaluate outcome quality, it does not include case volume activity. Unit performance provides this information, and it integrates quality and quantity into a single value. Congenital heart disease, Surgical performance, Quality assessment, Outcome INTRODUCTION Quality assessment of congenital heart surgery relies heavily on achieved early postoperative survival. This has to be adjusted according to the complexity of the operations performed and estimated in conformity with established mortality scores. Unadjusted mortality data would otherwise penalize centres that mostly manage high-risk procedures. The Society of Thoracic Surgeons (STS) and the European Association for Cardio-Thoracic Surgery (EACTS) Congenital Heart Surgery Mortality Score (STS-EACTS score), reported by O’Brien et al. [1] in the year 2009, is currently recognized as a sound instrument to assess and compare postoperative hospital survival in North American and European paediatric heart centres, but it does not take into account the number of procedures that are carried out. Case volume should be considered to also measure the workload achieved by the surgical team. Both quality and quantity should be evaluated to estimate performance in congenital heart surgery. In their introduction of the Aristotle complexity score, Lacour-Gayet et al. [2] defined surgical performance with the following formula: complexity × hospital survival = operative performance. The volume of procedures reported by participating centres was represented in the graph by 3 bullas of different sizes. A concept combining the evaluation of both quality and quantity was proposed for the 1st time in 2011 by Arenz et al. [3]: surgical case volume performance or unit performance. It was described as the product of surgical performance multiplied by the number of procedures carried out in a given time interval. This article analyses the outcome of our surgical practice according to the STS-EACTS score and to the notion of surgical performance. It applies the concept of unit performance, with particular comparison between achieved and expected unit performance, and discusses its possible impact on monitoring the activity and the calculated efficiency of different surgical paediatric heart centres. PATIENTS AND METHODS The study covers the years 2012–2016. All primary (main) procedures were considered except for closure of patent ductus arteriosus in premature newborns and procedures not recorded in the STS-EACTS score [1], such as diaphragm plication, extracorporeal membrane oxygenation and the removal of sternal wires. The Aristotle basic complexity scores attributed to main procedures were derived from the centre source data of the European Congenital Heart Surgeons Association (ECHSA) congenital database to which we sent anonymous parameters and outcomes of our patients. These are verified annually by the association coordinator. The STS-EACTS model-based mortality risks were used to estimate expected hospital mortality and, therefore, presumed survival for each procedure. Their 95% confidence interval (CI) enabled, accordingly, the calculation of the highest and lowest anticipated survival. The 95% CI was estimated according to the Bayesian credible interval [1]. Expected overall mortality rate is the sum of the predicted probabilities of postoperative death after all primary (main) procedures divided by the total number of procedures carried out in a given interval. Early postoperative deaths are those that occurred within a period of 30 days after surgery. Ratios of observed mortality divided by expected mortality (O/E), observed and expected operative performances and unit performances were calculated separately for each of the years 2012–2016. Results were also compared with those of the most recent STS database [4]. The χ2 test was used for group comparison. A P-value ≤0.05 was determined as significant. Means are given with standard deviation. The GraphPad Prism software (GraphPad Software, Inc., San Diego, CA, USA) was used for statistical calculations, computing equations and drawing graphs. Several regression equations were tested, and the regression function with the highest goodness of fit r2 was retained. All patients, their parents or legal guardians gave written informed consent prior to surgery to use anonymous data for external quality control and research. RESULTS In total, 2435 primary procedures were analysed, with a mean number per year of 487 ± 67 procedures. The most frequent operation was patch repair of ventricular septal defect (n = 213) with 100% hospital survival. One death occurred after arterial switch operation, and the 4th most frequent operation which was carried out 84 times. Survival was 88.6% (39/44) after the Norwood procedure. Table 1 lists details of postoperative mortality for the procedures that were carried out more than 10 times and their Aristotle basic complexity scores. Table 1: Procedures, expected and observed postoperative mortality Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  ABC: Aristotle basic complexity; ASD: atrial septal defect; ASO: arterial switch operation; AVSD: atrioventricular septal defect; DORV: double outlet right ventricle; HLHS: hypoplastic left heart syndrome; ICD: implantable cardioverter defibrillator; MBTS: modified Blalock–Taussig shunt; PA: pulmonary artery; PAPVC: partial anomalous pulmonary venous connection; RV: right ventricle; RVOT: right ventricular outflow tract; TAPVC: totally anomalous pulmonary venous connection; TCPC: total cavopulmonary connection; TOF: tetralogy of Fallot; VSD: ventricular septal defect. Table 1: Procedures, expected and observed postoperative mortality Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  ABC: Aristotle basic complexity; ASD: atrial septal defect; ASO: arterial switch operation; AVSD: atrioventricular septal defect; DORV: double outlet right ventricle; HLHS: hypoplastic left heart syndrome; ICD: implantable cardioverter defibrillator; MBTS: modified Blalock–Taussig shunt; PA: pulmonary artery; PAPVC: partial anomalous pulmonary venous connection; RV: right ventricle; RVOT: right ventricular outflow tract; TAPVC: totally anomalous pulmonary venous connection; TCPC: total cavopulmonary connection; TOF: tetralogy of Fallot; VSD: ventricular septal defect. One hundred and two deaths (95% CI 71–135 deaths) were predicted to occur, resulting in an expected survival of 95.8% (2333/2435) (95% CI 94.5–97.1%). Forty-three patients died after operation, with an observed survival of 98.2% (2392/2435). Overall O/E mortality ratio was, therefore, 0.42 (43/102). Difference between observed and expected mortality was highly significant: 1.8% vs 4.2% (P-value <0.0001). As listed in Table 2, this difference involved each of the 5 STS-EACTS mortality categories, with O/E ratios ranging from 0.25 (Category 1) to 0.51 (Category 4) and was in particular highly significant (P-value <0.01) for Categories 2, 4 and 5. Further comparison with the recent STS data [4] still concludes significantly lower mortality in this series: 1.8% vs 2.7%: P = 0.0048. In particular, mortality difference concerns the procedures in the STS-EACTS mortality Category 5. Table 2: STS-EACTS mortality categories and comparison of observed mortalities Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  EACTS: European Association of Cardio-Thoracic Surgery; STS: Society of Thoracic Surgeons. Table 2: STS-EACTS mortality categories and comparison of observed mortalities Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  EACTS: European Association of Cardio-Thoracic Surgery; STS: Society of Thoracic Surgeons. The mean Aristotle basic complexity score for the whole series attained 7.32 ± 2.52 points. Mean achieved surgical performance reached 7.19 ± 2.47 points. This was higher than 7.01 (95% CI 6.92–7.11), the mean expected surgical performance. Table 3 shows the evolution of O/E mortality ratios as well as observed and expected performances per year. The O/E ratio was the highest in 2013 (0.59) and the lowest in 2014 (0.23). However, achieved operative performance was the highest in 2012 (7.28 ± 2.54) and the lowest in 2014 (7.04 ± 2.52) as depicted in Fig. 1. Table 3: Evolution over the years of O/E ratios and of surgical and unit performances Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  ABC: Aristotle basic complexity; CI: confidence interval; O/E: observed mortality divided by expected mortality. Table 3: Evolution over the years of O/E ratios and of surgical and unit performances Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  ABC: Aristotle basic complexity; CI: confidence interval; O/E: observed mortality divided by expected mortality. Figure 1: View largeDownload slide Evolution of expected and observed surgical performance from years 2012 to 2016. Achieved operative performance was the highest in year 2012 (7.28 ± 2.54) and lowest in year 2014 (7.04 ± 2.52). Figure 1: View largeDownload slide Evolution of expected and observed surgical performance from years 2012 to 2016. Achieved operative performance was the highest in year 2012 (7.28 ± 2.54) and lowest in year 2014 (7.04 ± 2.52). Mean yearly attained unit performance was 3499 ± 461 points. It exceeded the upper limit of 95% CI of mean predicted unit performance: 3414 ± 445 points (95% CI 3364–3458 points). Table 3 shows how observed and expected unit performances evolved over the years. The highest figure of achieved unit performance was observed in 2016: 3980 points. The observed unit performances were not only higher than the expected performances every year, but these also exceeded the highest presumed performances. The computed best-fit non-linear regression lines graphing this evolution are sigmoidal (Fig. 2). Figure 2: View largeDownload slide Computed best-fit nonlinear regression graphing the evolution of expected and achieved unit performances over the years. The lines are sigmoidal. Related goodness of fit (r2) is displayed. The calculated maximal (top) values are 3715 and 3813 points for the expected lowest and highest unit performance, respectively, and 3863 points for the achieved performance. Figure 2: View largeDownload slide Computed best-fit nonlinear regression graphing the evolution of expected and achieved unit performances over the years. The lines are sigmoidal. Related goodness of fit (r2) is displayed. The calculated maximal (top) values are 3715 and 3813 points for the expected lowest and highest unit performance, respectively, and 3863 points for the achieved performance. DISCUSSION Model calibration can first be assessed by calculating the O/E mortality ratio. An O/E ratio above 1 indicates underestimation of mortality, and an O/E ratio below 1 signifies overestimation of mortality. Thus, with an O/E ratio of equal to 0.42 in this study and lower than 0.6 in each of the 5 mortality categories, the STS-EACTS score overestimated postoperative mortality and was not well calibrated to our surgical practice. This score was based on data provided by North American and European centres for the years 2002–2007. O’Brien et al. reported that, in the validation cohort of years 2007–2008, observed mortality rate was lower, ‘reflecting a trend towards lower mortality in a more contemporary sample’. The 2017 STS update outcome evaluation [4] also reports a mortality lower than expected by the STS-EACTS score for all 5 mortality categories. Our experience appears to confirm this tendency. If this is substantiated by further results from other centres and consequently by a multi-institutional study, the STS-EACTS score should be revised and recalibrated to match the current congenital heart surgical practice. However, it is to be noted that, as shown in Table 2, the STS-EACTS categories discriminated well between high-risk and low-risk procedures. Our findings might simply indicate that our centre performed better than the aggregate data used to estimate STS-EACTS mortality score. The Aristotle complexity score is the sole model to quantify surgical performance. It includes complexity and survival. Lacour-Gayet et al. [2] published basic surgical performances ranging from 5.67 to 6.90 points, mean 6.3 ± 0.4 points, for the period 1999–2003. Their values vary between 5.65 and 7.86 in more recent publications [3, 5–7]. In this series, they were the highest in 2012 (7.28) and the lowest in 2014 (7.04). One might, therefore, conclude that our department performed most effectively in 2012 and least efficiently in 2014. However, performance results appear different when quantity (case volume) is also considered. Indeed, figures of achieved unit performance indicate that the year 2014 (3795 points) was better than the year 2012 (2868 points). High figures of operative performance certainly attest remarkable quality of congenital heart surgery programmes, but quality alone does not suffice. Volume workload (intensity of activity) has also to be considered. By coupling quality and quantity, unit performance completes outcome assessment. In the publication by Arenz et al. [3] that introduced the concept of unit performance, basic unit performances ranged from 3036 to 3793 points, but expected performances were not mentioned. This study is the first to report anticipated unit performances. It clearly shows that our centre achieved a higher unit performance than expected based on the STS-EACTS mortality score. In conformity with Fig. 2, maximal unit performance seemed to have been reached. Our efficiency appeared to be maximized in our current setting. Consequently, one may argue that it could be increased only by obtaining supplementary human and/or material resources. The use of unit performance could help in the ongoing debate to define a benchmark for sustainable centres to ensure high quality. An institution managing a small number of procedures can achieve excellent results but often with large CIs. Centres dealing with a higher volume of cases usually perform better for lesions with a higher complexity [8–10]. In the analysis of verified data of EACTS Congenital Database, Kansy et al. [11] found that higher programmatic volume was associated with lower rates of mortality and morbidity. The small- and medium-volume centres had higher rates of major complications, and when complications occurred, the chance of rescue was higher in large-volume centres. There are units in Europe where the number of cases per year is very low and those in which results may appear unacceptable when compared with the current STS-EACTS score. Concern has been raised that constantly pushing towards the top performances could be counterproductive as there are a number of centres that struggle to reach the present expected standards. This underlines the importance of defining minimal requirements with agreed CIs. The Congenital Heart Disease Committee of the European Association for Cardio-Thoracic Surgery set a minimal number of 250 procedures per year [12]. However, no minimal surgical performance was quoted, as if the sole ‘sufficiently high’ volume would guarantee better quality outcome. It might be more judicious for decision makers to elaborate both minimal number of procedures and minimal operative performance: actually required threshold of unit performance for surgical paediatric centres, to qualify. For example, if the mean basic surgical performance of 6.3 (Lacour-Gayet et al. [2]) was chosen, the threshold of 1575 points (250 × 6.3) for unit performance might be contemplated as a reference. The potential for postoperative morbidity is the 2nd component of the Aristotle complexity score. Unit performance accommodates, therefore, potential for postoperative morbidity. This is somewhat a limitation in global quality assessment. Morbidity should be quantified in relation to observed complications as pioneered by Sata et al. [13] and compared with expected morbidity as in the model presented by Jacobs et al. [14]. Currently, there is no concept (a single value) which combines and estimates both mortality and morbidity. CONCLUSION In conclusion, the lower mortality than expected reported in this series suggests that the current STS-EACTS score needs recalibration and reappraisal to conform to the improved outcome in congenital heart surgery. Although surgical performance constitutes an excellent tool to assess and monitor an institution over time, it is not sufficient, as case volume activity is not included. Unit performance provides this supplementary information, and it integrates quality and quantity into a single value. It accurately quantified performance evolution of our unit over the 5-year period. This study constitutes a quality control for a single institution. The findings, especially the probable indication for STS-EACTS score adjustment, need confirmation from a multi-institutional evaluation. ACKNOWLEDGEMENTS We thank Anne Gale of the German Heart Center Berlin for editorial assistance. Conflict of interest: none declared. REFERENCES 1 O'Brien SM, Clarke DR, Jacobs JP, Jacobs ML, Lacour-Gayet FG, Pizarro C. An empirically based tool for analyzing mortality associated with congenital heart surgery. J Thorac Cardiovasc Surg  2009; 138: 1139– 53. Google Scholar CrossRef Search ADS PubMed  2 Lacour-Gayet F, Clarke D, Jacobs J, Comas J, Daebritz S, Daenen W et al.   The Aristotle score: a complexity-adjusted method to evaluate surgical results. Eur J Cardiothorac Surg  2004; 25: 911– 24. Google Scholar CrossRef Search ADS PubMed  3 Arenz C, Asfour B, Hraska V, Photiadis J, Haun C, Schindler E et al.   Congenital heart surgery: surgical performance according to the Aristotle complexity score. Eur J Cardiothorac Surg  2011; 39: e33– 7. Google Scholar CrossRef Search ADS PubMed  4 Jacobs JP, Mayer JE, Mavroudis C, O’Brien SM, Austin EH, Pasquali SK et al.   The Society of Thoracic Surgeons congenital heart surgery database: 2017 update on outcomes and quality. Ann Thorac Surg  2017; 103: 699– 709. Google Scholar CrossRef Search ADS PubMed  5 DeCampli WM, Burke RP. Interinstitutional comparison of risk-adjusted mortality and length of stay in congenital heart surgery. Ann Thorac Surg  2009; 88: 151– 6. Google Scholar CrossRef Search ADS PubMed  6 Vasdev S, Chauhan S, Malik M, Talwar S, Velayoudham D, Kiran U. Congenital heart surgery outcome analysis: Indian experience. Asian Cardiovasc Thorac Ann  2013; 21: 675– 82. Google Scholar CrossRef Search ADS PubMed  7 Joshi SS, Anthony G, Manasa D, Ashwini T, Jagadeesh AM, Borde DP et al.   Predicting mortality after congenital heart surgeries: evaluation of the Aristotle and Risk Adjustment in Congenital Heart Surgery-1 risk prediction scoring systems: a retrospective single center analysis of 1150 patients. Ann Card Anaesth  2014; 17: 266– 70. Google Scholar CrossRef Search ADS PubMed  8 Hannan EL, Racz M, Kavey RE, Quaegebeur JM, Williams R. Pediatric cardiac surgery: the effect of hospital and surgeon volume on in-hospital mortality. Pediatrics  1998; 101: 963– 9. Google Scholar CrossRef Search ADS PubMed  9 Hirsch JC, Gurney JG, Donohue JE, Gebremariam A, Bove EL, Ohye RG. Hospital mortality for Norwood and arterial switch operations as a function of institutional volume. Pediatr Cardiol  2008; 29: 713– 05. Google Scholar CrossRef Search ADS PubMed  10 Welke KF, O'Brien SM, Peterson ED, Ungerleider RM, Jacobs ML, Jacobs JP. The complex relationship between pediatric cardiac surgical case volumes and mortality rates in a national database. J Thorac Cardiovasc Surg  2009; 137: 1133– 40. Google Scholar CrossRef Search ADS PubMed  11 Kansy A, Ebels T, Schreiber C, Tobota Z, Maruszewski B. Association of center volume with outcomes: analysis of verified data of European Association for cardio Cardio-Thoracic Surgery Congenital Database. Ann Thorac Surg  2014; 98: 2159– 64. Google Scholar CrossRef Search ADS PubMed  12 Daenen W, Lacour GF, Aberg T, Comas JV, Daebritz SH, Di DR; EACTS Congenital Heart Disease Committee. Optimal structure of a congenital heart surgery Department in Europe. Eur J Cardiothorac Surg  2003; 24: 343– 51. Google Scholar CrossRef Search ADS PubMed  13 Sata S, Haun C, Weber T, Arenz C, Photiadis J, Hraska V et al.   A morbidity score for congenital heart surgery based on observed complications. Eur J Cardiothorac Surg  2012; 41: 898– 904. Google Scholar CrossRef Search ADS PubMed  14 Jacobs ML, O’Brien SM, Jacobs JP, Mavroudis C, Lacour-Gayet F, Pasquali SK et al.   An empirically based tool for analyzing morbidity associated with operations for congenital heart disease. J Thorac Cardiovasc Surg  2013; 145: 1046– 57. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Interactive CardioVascular and Thoracic Surgery Oxford University Press

Loading next page...
 
/lp/ou_press/assessment-of-a-congenital-heart-surgery-programme-a-reappraisal-wJI30U01Af
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1569-9293
eISSN
1569-9285
D.O.I.
10.1093/icvts/ivy090
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES To assess our practice according to the Society of Thoracic Surgeons and the European Association for Cardio-Thoracic Surgery (STS-EACTS) Mortality Score and to the new concept of unit performance. METHODS All main procedures carried out in the years 2012–2016 were analysed. The STS-EACTS model-based mortality risk procedure was used to calculate expected mortality. Surgical performance was estimated as the Aristotle complexity score multiplied by hospital survival. Unit performance was defined as surgical performance multiplied by the number of procedures. RESULTS In total, 2435 procedures were analysed. One hundred and two deaths (95% confidence interval 71–135 deaths) were expected; 43 patients died after operation. Observed mortality divided by expected mortality was 0.42. The ratio ranged from 0.23 (year 2014) to 0.59 (year 2013) and was <0.6 in all STS-EACTS mortality categories. The difference between observed and expected mortality was highly significant: 1.8% vs 4.2% (P-value <0.0001). Observed surgical and unit performances were, higher than expected performances every year. Achieved surgical performance was the highest in year 2012 (7.28 ± 2.54) and the lowest in year 2014 (7.04 ± 2.52). The highest figure of unit performance was achieved in year 2016: 3980 points. CONCLUSIONS The STS-EACTS score, currently recognized as a sound instrument to assess mortality after congenital heart surgery, overestimated postoperative mortality. If these results are confirmed by other centres, the model should be recalibrated to match the current surgical practice. Although surgical performance can evaluate outcome quality, it does not include case volume activity. Unit performance provides this information, and it integrates quality and quantity into a single value. Congenital heart disease, Surgical performance, Quality assessment, Outcome INTRODUCTION Quality assessment of congenital heart surgery relies heavily on achieved early postoperative survival. This has to be adjusted according to the complexity of the operations performed and estimated in conformity with established mortality scores. Unadjusted mortality data would otherwise penalize centres that mostly manage high-risk procedures. The Society of Thoracic Surgeons (STS) and the European Association for Cardio-Thoracic Surgery (EACTS) Congenital Heart Surgery Mortality Score (STS-EACTS score), reported by O’Brien et al. [1] in the year 2009, is currently recognized as a sound instrument to assess and compare postoperative hospital survival in North American and European paediatric heart centres, but it does not take into account the number of procedures that are carried out. Case volume should be considered to also measure the workload achieved by the surgical team. Both quality and quantity should be evaluated to estimate performance in congenital heart surgery. In their introduction of the Aristotle complexity score, Lacour-Gayet et al. [2] defined surgical performance with the following formula: complexity × hospital survival = operative performance. The volume of procedures reported by participating centres was represented in the graph by 3 bullas of different sizes. A concept combining the evaluation of both quality and quantity was proposed for the 1st time in 2011 by Arenz et al. [3]: surgical case volume performance or unit performance. It was described as the product of surgical performance multiplied by the number of procedures carried out in a given time interval. This article analyses the outcome of our surgical practice according to the STS-EACTS score and to the notion of surgical performance. It applies the concept of unit performance, with particular comparison between achieved and expected unit performance, and discusses its possible impact on monitoring the activity and the calculated efficiency of different surgical paediatric heart centres. PATIENTS AND METHODS The study covers the years 2012–2016. All primary (main) procedures were considered except for closure of patent ductus arteriosus in premature newborns and procedures not recorded in the STS-EACTS score [1], such as diaphragm plication, extracorporeal membrane oxygenation and the removal of sternal wires. The Aristotle basic complexity scores attributed to main procedures were derived from the centre source data of the European Congenital Heart Surgeons Association (ECHSA) congenital database to which we sent anonymous parameters and outcomes of our patients. These are verified annually by the association coordinator. The STS-EACTS model-based mortality risks were used to estimate expected hospital mortality and, therefore, presumed survival for each procedure. Their 95% confidence interval (CI) enabled, accordingly, the calculation of the highest and lowest anticipated survival. The 95% CI was estimated according to the Bayesian credible interval [1]. Expected overall mortality rate is the sum of the predicted probabilities of postoperative death after all primary (main) procedures divided by the total number of procedures carried out in a given interval. Early postoperative deaths are those that occurred within a period of 30 days after surgery. Ratios of observed mortality divided by expected mortality (O/E), observed and expected operative performances and unit performances were calculated separately for each of the years 2012–2016. Results were also compared with those of the most recent STS database [4]. The χ2 test was used for group comparison. A P-value ≤0.05 was determined as significant. Means are given with standard deviation. The GraphPad Prism software (GraphPad Software, Inc., San Diego, CA, USA) was used for statistical calculations, computing equations and drawing graphs. Several regression equations were tested, and the regression function with the highest goodness of fit r2 was retained. All patients, their parents or legal guardians gave written informed consent prior to surgery to use anonymous data for external quality control and research. RESULTS In total, 2435 primary procedures were analysed, with a mean number per year of 487 ± 67 procedures. The most frequent operation was patch repair of ventricular septal defect (n = 213) with 100% hospital survival. One death occurred after arterial switch operation, and the 4th most frequent operation which was carried out 84 times. Survival was 88.6% (39/44) after the Norwood procedure. Table 1 lists details of postoperative mortality for the procedures that were carried out more than 10 times and their Aristotle basic complexity scores. Table 1: Procedures, expected and observed postoperative mortality Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  ABC: Aristotle basic complexity; ASD: atrial septal defect; ASO: arterial switch operation; AVSD: atrioventricular septal defect; DORV: double outlet right ventricle; HLHS: hypoplastic left heart syndrome; ICD: implantable cardioverter defibrillator; MBTS: modified Blalock–Taussig shunt; PA: pulmonary artery; PAPVC: partial anomalous pulmonary venous connection; RV: right ventricle; RVOT: right ventricular outflow tract; TAPVC: totally anomalous pulmonary venous connection; TCPC: total cavopulmonary connection; TOF: tetralogy of Fallot; VSD: ventricular septal defect. Table 1: Procedures, expected and observed postoperative mortality Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  Procedure  Number  ABC score [2]  Predicted deaths [1]  Observed deaths  VSD repair, patch  213  6.0  1.92  0  ASD repair, patch  141  3.0  0.42  0  Bidirectional cavopulmonary anastomosis  96  7.0  2.59  0  Arterial switch operation  84  10.0  4.03  1  Valvuloplasty, aortic  78  8.0  1.48  0  TOF repair, ventriculotomy, transannular patch  75  8.0  2.03  1  Coarctation repair, end to end, extended  71  8.0  1.78  0  Complete AVSD repair  70  9.0  3.22  0  Conduit placement, RV to PA  68  7.5  4.56  2  PA banding  57  6.0  5.59  3  Conduit reoperation  55  8.0  0.77  0  Valve replacement, pulmonic  54  6.5  0.70  0  Valvuloplasty, mitral  53  8.0  1.01  0  PAPVC repair  52  5.0  0.78  0  Pacemaker implantation, permanent  51  3.0  1.12  1  Fontan, TCPC, external conduit, fenestrated  47  9.0  1.41  0  Norwood procedure  44  14.5  10.38  5  Aortic stenosis, subvalvular, repair, with myectomy  41  6.3  0.25  1  Aortic valve replacement, mechanical  35  7.0  0.60  0  Pacemaker procedure  34  3.0  0.48  0  Ross–Konno procedure  33  12.5  3.10  1  Ebstein's repair  32  10.0  3.04  0  Aortic stenosis, subvalvular, repair  31  6.3  0.19  0  ASD repair, patch + PAPVC repair  31  5.0  0.19  0  MBTS  31  6.3  2.76  2  PAVSD repair  30  4.0  0.15  0  Aortic arch repair  29  7.0  2.26  2  ASO + VSD repair  29  11.0  2.38  0  TOF repair ventriculotomy, non-transannular patch  29  7.5  0.44  0  Valvuloplasty, tricuspid  29  7.0  1.07  0  ASD repair, primary closure  27  3.0  0.24  0  Fontan TCPC extracard, non-fenestrated  27  9.0  0.86  1  TOF, no ventriculotomy repair  26  8.0  0.39  1  TAPVC repair  25  9.0  2.80  2  Damus–Kaye–Stansel procedure  24  9.5  4.10  1  Mitral valve replacement  24  7.5  1.75  0  Vascular ring repair  23  6.0  0.21  0  AVSD repair, intermediate  22  5.0  0.35  0  Ross procedure  21  10.3  0.46  0  Shunt, systemic to pulmonary, central  21  6.8  2.54  3  HLHS biventricular repair  18  15.0  1.96  0  PA reconstruction (plasty), main (trunk)  18  6.0  0.34  0  Aortic aneurysm repair  16  8.8  0.42  1  Aortic root replacement, valve sparing  16  8.5  0.54  0  Aortic stenosis, supravalvular, repair  15  5.5  0.42  0  DORV, intraventricular tunnel  15  10.3  1.20  0  Valvuloplasty, pulmonic  15  5.6  0.41  0  PA reconstruction, branch within hilar bifurcation  14  7.8  0.60  1  Truncus arteriosus repair  14  11.0  1.97  1  ICD  12  4.0  0.12  0  Pericardial drainage procedure  12  3.0  0.90  1  Unidirectional Glenn  12  7.0  0.18  0  VSD repair, primary closure  12  6.0  0.14  0  Aortic arch repair + VSD repair  11  10.0  1.08  0  Aortic root replacement, mechanical  10  8.8  0.24  0  RVOT procedure  10  6.5  0.26  0  Tricuspid valve replacement  10  7.5  0.38  1  Others  242  8.6 ± 2.4  16.66  11  Total  2435  7.32 ± 2.52  102.21 (4.2%)  43 (1.8%)  ABC: Aristotle basic complexity; ASD: atrial septal defect; ASO: arterial switch operation; AVSD: atrioventricular septal defect; DORV: double outlet right ventricle; HLHS: hypoplastic left heart syndrome; ICD: implantable cardioverter defibrillator; MBTS: modified Blalock–Taussig shunt; PA: pulmonary artery; PAPVC: partial anomalous pulmonary venous connection; RV: right ventricle; RVOT: right ventricular outflow tract; TAPVC: totally anomalous pulmonary venous connection; TCPC: total cavopulmonary connection; TOF: tetralogy of Fallot; VSD: ventricular septal defect. One hundred and two deaths (95% CI 71–135 deaths) were predicted to occur, resulting in an expected survival of 95.8% (2333/2435) (95% CI 94.5–97.1%). Forty-three patients died after operation, with an observed survival of 98.2% (2392/2435). Overall O/E mortality ratio was, therefore, 0.42 (43/102). Difference between observed and expected mortality was highly significant: 1.8% vs 4.2% (P-value <0.0001). As listed in Table 2, this difference involved each of the 5 STS-EACTS mortality categories, with O/E ratios ranging from 0.25 (Category 1) to 0.51 (Category 4) and was in particular highly significant (P-value <0.01) for Categories 2, 4 and 5. Further comparison with the recent STS data [4] still concludes significantly lower mortality in this series: 1.8% vs 2.7%: P = 0.0048. In particular, mortality difference concerns the procedures in the STS-EACTS mortality Category 5. Table 2: STS-EACTS mortality categories and comparison of observed mortalities Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  EACTS: European Association of Cardio-Thoracic Surgery; STS: Society of Thoracic Surgeons. Table 2: STS-EACTS mortality categories and comparison of observed mortalities Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  Category  STS-EACTS [1] expected mortality (%)  This series observed mortality   STS: 2012–2015 [4] observed mortality   P-value   Cases  Deaths (%)  Cases  Deaths (%)  This series versus STS-EACTS  This series versus STS  1  0.8  913  2 (0.2)  28 896  144 (0.5)  0.07  0.34  2  2.6  716  6 (0.8)  35 519  604 (1.7)  0.005  0.10  3  5.0  282  6 (2.1)  10 629  276 (2.6)  0.04  0.76  4  9.9  443  23 (5.0)  19 038  1314 (6.9)  0.001  0.13  5  23.1  81  6 (7.4)  3753  604 (16.1)  0.001  0.049  Total  4.2 (2.9–5.5)  2435  43 (1.8)  97 835  2666 (2.7)  <0.0001  0.0048  EACTS: European Association of Cardio-Thoracic Surgery; STS: Society of Thoracic Surgeons. The mean Aristotle basic complexity score for the whole series attained 7.32 ± 2.52 points. Mean achieved surgical performance reached 7.19 ± 2.47 points. This was higher than 7.01 (95% CI 6.92–7.11), the mean expected surgical performance. Table 3 shows the evolution of O/E mortality ratios as well as observed and expected performances per year. The O/E ratio was the highest in 2013 (0.59) and the lowest in 2014 (0.23). However, achieved operative performance was the highest in 2012 (7.28 ± 2.54) and the lowest in 2014 (7.04 ± 2.52) as depicted in Fig. 1. Table 3: Evolution over the years of O/E ratios and of surgical and unit performances Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  ABC: Aristotle basic complexity; CI: confidence interval; O/E: observed mortality divided by expected mortality. Table 3: Evolution over the years of O/E ratios and of surgical and unit performances Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  Parameters  Year 2012  Year 2013  Year 2014  Year 2015  Year 2016  Total/mean  Number of procedures  394  441  539  509  552  2435  ABC score  7.39 ± 2.58  7.38 ± 2.4  7.1 ± 2.54  7.33 ± 2.54  7.41 ± 2.52  7.32 ± 2.52  Observed deaths  6  10  5  8  14  43  Observed mortality (%)  1.5  2.3  0.9  1.6  2.5  1.8  Expected deaths  16  17  22  21  26  102  Expected mortality (%)  4.1  3.9  4.1  4.1  4.7  4.2  95% CI  3.0–5.6  2.6–5.4  2.7–5.7  3.0–5.8  3.3–5.4  2.9–5.5  O/E ratio  0.38  0.59  0.23  0.38  0.54  0.42  Observed surgical performance  7.28 ± 2.54  7.21 ± 2.34  7.04 ± 2.52  7.21 ± 2.50  7.22 ± 2.48  7.19 ± 2.47  Expected surgical performance (95% CI)  7.09 (6.98–7.17)  7.1 (6.98–7.19)  6.82 (6.70–6.91)  7.02 (6.90–7.11)  7.06 (7.01–7.16)  7.01 (6.92–7.11)  Observed unit performance  2868  3180  3795  3670  3980  17 493  Expected unit performance (95% CI)  2793 (2750–2825)  3131 (3078–3171)  3676 (3611–3724)  3573 (3512–3619)  3897 (3870–3952)  17 070 (16 821–17 291)  ABC: Aristotle basic complexity; CI: confidence interval; O/E: observed mortality divided by expected mortality. Figure 1: View largeDownload slide Evolution of expected and observed surgical performance from years 2012 to 2016. Achieved operative performance was the highest in year 2012 (7.28 ± 2.54) and lowest in year 2014 (7.04 ± 2.52). Figure 1: View largeDownload slide Evolution of expected and observed surgical performance from years 2012 to 2016. Achieved operative performance was the highest in year 2012 (7.28 ± 2.54) and lowest in year 2014 (7.04 ± 2.52). Mean yearly attained unit performance was 3499 ± 461 points. It exceeded the upper limit of 95% CI of mean predicted unit performance: 3414 ± 445 points (95% CI 3364–3458 points). Table 3 shows how observed and expected unit performances evolved over the years. The highest figure of achieved unit performance was observed in 2016: 3980 points. The observed unit performances were not only higher than the expected performances every year, but these also exceeded the highest presumed performances. The computed best-fit non-linear regression lines graphing this evolution are sigmoidal (Fig. 2). Figure 2: View largeDownload slide Computed best-fit nonlinear regression graphing the evolution of expected and achieved unit performances over the years. The lines are sigmoidal. Related goodness of fit (r2) is displayed. The calculated maximal (top) values are 3715 and 3813 points for the expected lowest and highest unit performance, respectively, and 3863 points for the achieved performance. Figure 2: View largeDownload slide Computed best-fit nonlinear regression graphing the evolution of expected and achieved unit performances over the years. The lines are sigmoidal. Related goodness of fit (r2) is displayed. The calculated maximal (top) values are 3715 and 3813 points for the expected lowest and highest unit performance, respectively, and 3863 points for the achieved performance. DISCUSSION Model calibration can first be assessed by calculating the O/E mortality ratio. An O/E ratio above 1 indicates underestimation of mortality, and an O/E ratio below 1 signifies overestimation of mortality. Thus, with an O/E ratio of equal to 0.42 in this study and lower than 0.6 in each of the 5 mortality categories, the STS-EACTS score overestimated postoperative mortality and was not well calibrated to our surgical practice. This score was based on data provided by North American and European centres for the years 2002–2007. O’Brien et al. reported that, in the validation cohort of years 2007–2008, observed mortality rate was lower, ‘reflecting a trend towards lower mortality in a more contemporary sample’. The 2017 STS update outcome evaluation [4] also reports a mortality lower than expected by the STS-EACTS score for all 5 mortality categories. Our experience appears to confirm this tendency. If this is substantiated by further results from other centres and consequently by a multi-institutional study, the STS-EACTS score should be revised and recalibrated to match the current congenital heart surgical practice. However, it is to be noted that, as shown in Table 2, the STS-EACTS categories discriminated well between high-risk and low-risk procedures. Our findings might simply indicate that our centre performed better than the aggregate data used to estimate STS-EACTS mortality score. The Aristotle complexity score is the sole model to quantify surgical performance. It includes complexity and survival. Lacour-Gayet et al. [2] published basic surgical performances ranging from 5.67 to 6.90 points, mean 6.3 ± 0.4 points, for the period 1999–2003. Their values vary between 5.65 and 7.86 in more recent publications [3, 5–7]. In this series, they were the highest in 2012 (7.28) and the lowest in 2014 (7.04). One might, therefore, conclude that our department performed most effectively in 2012 and least efficiently in 2014. However, performance results appear different when quantity (case volume) is also considered. Indeed, figures of achieved unit performance indicate that the year 2014 (3795 points) was better than the year 2012 (2868 points). High figures of operative performance certainly attest remarkable quality of congenital heart surgery programmes, but quality alone does not suffice. Volume workload (intensity of activity) has also to be considered. By coupling quality and quantity, unit performance completes outcome assessment. In the publication by Arenz et al. [3] that introduced the concept of unit performance, basic unit performances ranged from 3036 to 3793 points, but expected performances were not mentioned. This study is the first to report anticipated unit performances. It clearly shows that our centre achieved a higher unit performance than expected based on the STS-EACTS mortality score. In conformity with Fig. 2, maximal unit performance seemed to have been reached. Our efficiency appeared to be maximized in our current setting. Consequently, one may argue that it could be increased only by obtaining supplementary human and/or material resources. The use of unit performance could help in the ongoing debate to define a benchmark for sustainable centres to ensure high quality. An institution managing a small number of procedures can achieve excellent results but often with large CIs. Centres dealing with a higher volume of cases usually perform better for lesions with a higher complexity [8–10]. In the analysis of verified data of EACTS Congenital Database, Kansy et al. [11] found that higher programmatic volume was associated with lower rates of mortality and morbidity. The small- and medium-volume centres had higher rates of major complications, and when complications occurred, the chance of rescue was higher in large-volume centres. There are units in Europe where the number of cases per year is very low and those in which results may appear unacceptable when compared with the current STS-EACTS score. Concern has been raised that constantly pushing towards the top performances could be counterproductive as there are a number of centres that struggle to reach the present expected standards. This underlines the importance of defining minimal requirements with agreed CIs. The Congenital Heart Disease Committee of the European Association for Cardio-Thoracic Surgery set a minimal number of 250 procedures per year [12]. However, no minimal surgical performance was quoted, as if the sole ‘sufficiently high’ volume would guarantee better quality outcome. It might be more judicious for decision makers to elaborate both minimal number of procedures and minimal operative performance: actually required threshold of unit performance for surgical paediatric centres, to qualify. For example, if the mean basic surgical performance of 6.3 (Lacour-Gayet et al. [2]) was chosen, the threshold of 1575 points (250 × 6.3) for unit performance might be contemplated as a reference. The potential for postoperative morbidity is the 2nd component of the Aristotle complexity score. Unit performance accommodates, therefore, potential for postoperative morbidity. This is somewhat a limitation in global quality assessment. Morbidity should be quantified in relation to observed complications as pioneered by Sata et al. [13] and compared with expected morbidity as in the model presented by Jacobs et al. [14]. Currently, there is no concept (a single value) which combines and estimates both mortality and morbidity. CONCLUSION In conclusion, the lower mortality than expected reported in this series suggests that the current STS-EACTS score needs recalibration and reappraisal to conform to the improved outcome in congenital heart surgery. Although surgical performance constitutes an excellent tool to assess and monitor an institution over time, it is not sufficient, as case volume activity is not included. Unit performance provides this supplementary information, and it integrates quality and quantity into a single value. It accurately quantified performance evolution of our unit over the 5-year period. This study constitutes a quality control for a single institution. The findings, especially the probable indication for STS-EACTS score adjustment, need confirmation from a multi-institutional evaluation. ACKNOWLEDGEMENTS We thank Anne Gale of the German Heart Center Berlin for editorial assistance. Conflict of interest: none declared. REFERENCES 1 O'Brien SM, Clarke DR, Jacobs JP, Jacobs ML, Lacour-Gayet FG, Pizarro C. An empirically based tool for analyzing mortality associated with congenital heart surgery. J Thorac Cardiovasc Surg  2009; 138: 1139– 53. Google Scholar CrossRef Search ADS PubMed  2 Lacour-Gayet F, Clarke D, Jacobs J, Comas J, Daebritz S, Daenen W et al.   The Aristotle score: a complexity-adjusted method to evaluate surgical results. Eur J Cardiothorac Surg  2004; 25: 911– 24. Google Scholar CrossRef Search ADS PubMed  3 Arenz C, Asfour B, Hraska V, Photiadis J, Haun C, Schindler E et al.   Congenital heart surgery: surgical performance according to the Aristotle complexity score. Eur J Cardiothorac Surg  2011; 39: e33– 7. Google Scholar CrossRef Search ADS PubMed  4 Jacobs JP, Mayer JE, Mavroudis C, O’Brien SM, Austin EH, Pasquali SK et al.   The Society of Thoracic Surgeons congenital heart surgery database: 2017 update on outcomes and quality. Ann Thorac Surg  2017; 103: 699– 709. Google Scholar CrossRef Search ADS PubMed  5 DeCampli WM, Burke RP. Interinstitutional comparison of risk-adjusted mortality and length of stay in congenital heart surgery. Ann Thorac Surg  2009; 88: 151– 6. Google Scholar CrossRef Search ADS PubMed  6 Vasdev S, Chauhan S, Malik M, Talwar S, Velayoudham D, Kiran U. Congenital heart surgery outcome analysis: Indian experience. Asian Cardiovasc Thorac Ann  2013; 21: 675– 82. Google Scholar CrossRef Search ADS PubMed  7 Joshi SS, Anthony G, Manasa D, Ashwini T, Jagadeesh AM, Borde DP et al.   Predicting mortality after congenital heart surgeries: evaluation of the Aristotle and Risk Adjustment in Congenital Heart Surgery-1 risk prediction scoring systems: a retrospective single center analysis of 1150 patients. Ann Card Anaesth  2014; 17: 266– 70. Google Scholar CrossRef Search ADS PubMed  8 Hannan EL, Racz M, Kavey RE, Quaegebeur JM, Williams R. Pediatric cardiac surgery: the effect of hospital and surgeon volume on in-hospital mortality. Pediatrics  1998; 101: 963– 9. Google Scholar CrossRef Search ADS PubMed  9 Hirsch JC, Gurney JG, Donohue JE, Gebremariam A, Bove EL, Ohye RG. Hospital mortality for Norwood and arterial switch operations as a function of institutional volume. Pediatr Cardiol  2008; 29: 713– 05. Google Scholar CrossRef Search ADS PubMed  10 Welke KF, O'Brien SM, Peterson ED, Ungerleider RM, Jacobs ML, Jacobs JP. The complex relationship between pediatric cardiac surgical case volumes and mortality rates in a national database. J Thorac Cardiovasc Surg  2009; 137: 1133– 40. Google Scholar CrossRef Search ADS PubMed  11 Kansy A, Ebels T, Schreiber C, Tobota Z, Maruszewski B. Association of center volume with outcomes: analysis of verified data of European Association for cardio Cardio-Thoracic Surgery Congenital Database. Ann Thorac Surg  2014; 98: 2159– 64. Google Scholar CrossRef Search ADS PubMed  12 Daenen W, Lacour GF, Aberg T, Comas JV, Daebritz SH, Di DR; EACTS Congenital Heart Disease Committee. Optimal structure of a congenital heart surgery Department in Europe. Eur J Cardiothorac Surg  2003; 24: 343– 51. Google Scholar CrossRef Search ADS PubMed  13 Sata S, Haun C, Weber T, Arenz C, Photiadis J, Hraska V et al.   A morbidity score for congenital heart surgery based on observed complications. Eur J Cardiothorac Surg  2012; 41: 898– 904. Google Scholar CrossRef Search ADS PubMed  14 Jacobs ML, O’Brien SM, Jacobs JP, Mavroudis C, Lacour-Gayet F, Pasquali SK et al.   An empirically based tool for analyzing morbidity associated with operations for congenital heart disease. J Thorac Cardiovasc Surg  2013; 145: 1046– 57. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Interactive CardioVascular and Thoracic SurgeryOxford University Press

Published: Mar 22, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off