Cover Page. The handle holds various files of this Leiden University dissertation.

Transcription

1 Cover Page The handle holds various files of this Leiden University dissertation. Author: Hoohenkerk, Gerard Joannes Franciscus Title: Surgical correction of atrioventricular septal defect Date:

2 More Than 30 Years Experience With Surgical Correction of Atrioventricular Septal Defects Gerard J. F. Hoohenkerk¹, MD, Eline F. Bruggemans¹, MS, Marry Rijlaarsdam², MD, Paul H. Schoof¹, MD, PhD, Dave R. Koolbergen¹, MD, PhD, and Mark G. Hazekamp¹, MD, PhD. Departments of Cardiothoracic Surgery¹, and Pediatric Cardiology², Leiden University Medical Center, Leiden, The Netherlands. Annals of Thoracic Surgery 2010;90: Pediatric Cardiac Surgery

3 Abstract Background. Outcome of surgical correction of atrioventricular septal defects (AVSD) still varies despite enhanced results. We reviewed our 30-year experience with AVSD repair and identified risk factors for mortality and reoperation. Methods. Between 1975 and 2006, 312 patients underwent surgery for complete AVSD (n = 209; 67.0%), partial AVSD (n = 76; 24.4%), or intermediate AVSD (n = 27; 8.6%). Mean age was 2.4 ± 3.9 years; 142 patients (45.5%) were younger than 6 months. Follow-up was 99.0% complete. Results. There were 26 in-hospital deaths (8.3%) and 6 late deaths (2.1% of 283). Estimated overall survival for the total study population was 91.3%, 90.6%, and 88.6% at 1, 5, and 15 years, respectively. In the multivariable logistic regression analysis, surgical era 1975 to 1995 (p < 0.001) and younger age (p = 0.004) were found to be independent risk factors for early mortality, whereas preoperative AV valve insufficiency showed a tendency toward statistical significance (p = 0.052). Of the hospital survivors, 43 patients required a late reoperation. Estimated freedom from late reoperation was 96.4%, 89.3%, and 81.8% at 1, 5, and 15 years, respectively. Multivariable Cox regression analysis showed associated cardiovascular anomalies (p < 0.001), left AV valve dysplasia (p < 0.001), and absence of cleft closure (p = 0.003) to be independent risk factors for late reoperation. Conclusions. AVSD repair can be accomplished with good long-term results. Early surgical era, associated cardiovascular anomalies, left AV valve dysplasia, and absence of cleft closure negatively influence survival and risk of reoperation. 24

4 Introduction Atrioventricular septal defect (AVSD) includes complete, partial, and intermediate AVSD. In complete AVSD (c-avsd), there is a common AV valve for both ventricles and interatrial and interventricular communication. In partial AVSD (p-avsd), also referred to as primum type atrial septal defect (ASD I), there are separate right and left AV valve orifices, and interventricular communication is lacking. An intermediate form of AVSD (i-avsd) is defined as having a scooped out interventricular septum with the AV valves being connected to the top of the septum by fibrous tissue curtains and tendinous chordae, consequently resulting in a small or absent VSD component. Most investigations that evaluate outcome of surgical correction of AVSD focus on either c- AVSD or p-avsd [1-6]. In this study, we evaluated whether type of AVSD (c-avsd, p- AVSD, or i-avsd), among other factors, is a risk factor for mortality and reoperation after surgical repair. In addition, we studied whether risk factors for mortality and reoperation differed for the three AVSD types. Hereto, we reviewed our institution s 30-year experience with surgical repair of AVSD. Patients and Methods Patient Population Between January 1975 and May 2006, 312 consecutive patients underwent surgical correction of AVSD at our institution. Complete AVSD was observed in 209 patients (67.0%), p-avsd in 76 (24.4%), and i-avsd in 27 (8.6%). Patients with associated cardiovascular anomalies, namely, tetralogy of Fallot (TOF [n = 24; 7.7%]), presence of an accessory orifice within the left-sided AV valve (double orifice [DO] left AV valve; n = 21; 6.7%), coarctation of aorta (n = 11; 3.5%), and left ventricular outflow tract obstruction (n = 2; 0.6%), were included. Three patients (1.0%) were diagnosed with combined DO left AV valve coarctation of aorta. All patients with unbalanced forms of AVSD or common AV valve malalignment were excluded, as were patients with other associated cardiovascular anomalies. Secundum type ASD and patent ductus arteriosus were not marked as associated cardiovascular anomalies. The study was approved with waiver of consent by the Ethics Committee of the institution. 25

5 Surgical Technique Six pediatric cardiac surgeons performed 297 of the 312 procedures, with very little variation in surgical technique over time. Some older patients with p-avsd were operated on by surgeons for adult patients. All operations were performed with cardiopulmonary bypass and moderate hypothermia. St. Thomas cardioplegia solution administered every 30 minutes was used throughout the years. Intraoperative transesophageal echocardiography both before and after repair was routinely used since Atrioventricular Septal Defects In patients with c-avsd, the septal defects were all closed using a two-patch technique but with different patch materials (autologous or heterologous pericardium, or Dacron [C.R. Bard, Haverhill, PA]). A one-patch technique was used in all p-avsd patients. In patients with i- AVSD, a two-patch technique was applied in 9 patients, and a one-patch technique served for repair in the other 18 patients. The cleft was closed by interrupted sutures to such an extent that regurgitation was managed optimally without creating a valvular stenosis. Cleft closure was performed in 180 of 209 c-avsd patients (86.1%), in 42 of 76 p-avsd patients (55.3%), and in 19 of 27 i-avsd patients (70.4%). Associated Cardiovascular Anomalies Tetralogy of Fallot, DO left AV valve, coarctation of aorta, and left ventricular outflow tract obstruction were all simultaneously repaired with the AVSD. In the 24 patients with c-avsd- TOF, right ventricular outflow tract enlargement was performed by a transatrial, transpulmonary approach. A transannular patch was used in 20 of 24 patients. The cleft was closed in 22 of 24 patients [7]. In the 21 patients with DO left AV valve, the accessory orifice 26

6 was repaired in 14 patients with suture (n = 9), patch (n = 3), or resection of the tissue bridge and division of the papillary muscle of the accessory orifice (n = 2) [8]. Data Collection and Follow-Up Patient data were obtained by reviewing both inpatient and outpatient medical records, including surgeon s notes, operative reports, hospital charts, and cardiac catheterization and echocardiographic reports. The majority of the patients underwent routine follow-up at our Department of Pediatric Cardiology. The closing interval for follow-up was February 2006 to December Follow-up data were complete for 283 of 286 hospital survivors (99.0%). For 7 patients, follow-up information was obtained by a telephone call to the patient s family or physician, or both. Three patients with c-avsd were lost to follow-up. The median followup period was 10.7 years (range, 0.4 to 29.3). For c-avsd patients (n = 184), median followup was 9.9 years (range, 0.4 to 28.9); for p-avsd patients (n = 73), it was 15.0 years (range, 1.9 to 29.3); and for i-avsd patients (n = 26), it was 6.2 years (range, 1.9 to 18.7). Definitions and Endpoints Left AV valve dysplasia was defined as a left AV valve with stiffened, thickened, and myxoid degenerated valve tissue and curling-in of the leaflets. Endpoints in the study were in-hospital and late mortality and early and late reoperation. In-hospital mortality and early reoperation were defined as death or reoperation before hospital discharge or within 30 days. Late mortality and reoperation were defined as all-cause death or reoperation more than 30 days after the surgical correction for AVSD. The AVSD repairs were divided into two surgical eras to assess the impact of surgical era on mortality and reoperation: 1975 to 1995, and 1996 to To assess the impact of concomitant procedures due to associated cardiovascular anomalies, associated cardiovascular anomalies were defined as absent or present because of the relatively small patient numbers for these anomalies in the three AVSD groups. Statistical Analysis Estimates of overall survival and freedom from late reoperation were obtained by means of the Kaplan- Meier method, with differences among the three AVSD groups being tested by the log rank test. Binary logistic regression analysis was used to examine the relationship between potential risk factors and in-hospital mortality. Separate logistic regression analyses were used first to assess the individual impact of each variable. Subsequently, a logistic regression analysis, using all variables in a stepwise method was performed to identify 27

7 independent risk factors. Cox proportional hazard models (for each variable separately and for all variables combined in a stepwise method) were performed to examine predictors for late reoperation. Potential risk factors tested in relation to in-hospital mortality and late reoperation were surgical era, type of AVSD, age, sex, Down syndrome, associated cardiovascular anomalies, left AV valve dysplasia, preoperative left AV valve insufficiency, and cleft closure. Probabilities used for entrance and removal in the stepwise method were 0.05 and 0.10, respectively. A p value of less than 0.05 (two-sided) was considered statistically significant. All statistical analyses were performed using the SPSS statistical software program for Windows, version (SPSS, Chicago, IL). Results Patient Characteristics Patient characteristics for the three AVSD groups and the total study population are given in Table 1. The mean age at the time of AVSD repair was 2.4 ± 3.9 years. Two hundred three repairs (65.1%) were performed in infants younger than 1 year, and 142 (45.5%) in infants younger than 6 months. Mean age at operation decreased over the last 3 decades from 5.2 years to 8.1 months (Fig 1). 28

8 29

9 Mortality There were 26 (8.3%) in-hospital deaths. In-hospital mortality for c-avsd patients was 22 of 209 (10.5%); for p-avsd patients, it was 3 of 76 (3.9%); and for i-avsd patients, it was 1 of 27 (3.7%). Causes of in-hospital mortality are listed in Table 2. In-hospital mortality decreased from 21 of 158 patients (13.3%) before January 1996 to 5 of 154 (3.2%) since January For c-avsd patients, it decreased from 18 of 87 patients (20.7%) to 4 of 122 (3.3%). In-hospital mortality among infants younger than 6 months was 16 of 142 (11.3%) compared with 10 of 170 (5.9%) among infants aged 6 months and older. Among infants younger than 6 months, in-hospital mortality decreased from 11 of 29 patients (37.9%) before January 1996 to 5 of 113 (4.4%) since January Late deaths occurred in 6 of 283 patients (2.1% [c- AVSD, n = 2; p-avsd, n = 1; i-avsd, n = 3]), of which 4 were cardiac related. Two patients died of late phase sepsis (1 c-avsd and 1 i-avsd patient), 1 of cardiac failure (i-avsd patient), and 1 of respiratory failure (i- AVSD patient). Two patients died in a traffic accident. Estimated overall survival for all AVSD patients (n = 309) was 91.3% at 1 year, 90.6% at 5 years, and 88.6% at 15 years. The log rank test showed no statistically significant difference in overall survival between c-avsd, p-avsd, and i-avsd patients (Fig 2). For c-avsd patients, it was 88.8% at 1 year, and 88.3% at 5 years and 15 years. 30

10 Estimated overall survival for p-avsd patients was 96.1% at 1 and 5 years, and 94.5% at 15 years. Estimated overall survival for i-avsd patients was 96.3% at 1 year and 92.4% at 5 years. Figures for i-avsd patients at 10 and 15 years could not be judged as reliable due to the small number of patients at risk. Predictors of In-Hospital Mortality For the total study population (n = 312), univariable logistic regression analyses revealed surgical era and age to be risk factors for in-hospital mortality (Table 3), with patients being operated on before 1996 being at greater risk (odds ratio [OR] 0.22, 95% confidence interval [CI]: 0.08 to 0.60, p = 0.003) as well as younger patients (OR 0.68, 95% CI: 0.47 to 0.99, p = 0.045). In the multivariable logistic regression analysis, surgical era (OR 0.10, 95% CI: 0.03 to 0.28, p < 0.001) and age (OR 0.45, 95% CI: 0.26 to 0.78, p = 0.004) were found to be independent risk factors, whereas preoperative left AV valve insufficiency showed a tendency toward statistical significance (p = 0.052; moderate AV valve insufficiency: OR 4.13, 95% CI: 1.24 to 13.80, p = 0.021; severe AV valve insufficiency: OR 4.04, 95% CI: 1.12 to 14.50, p = 0.032). 31

11 32

12 Similar analyses were performed for the subgroup of c -AVSD patients. The relatively small numbers in the other two subgroups did not allow for separate analyses. Risk factors for inhospital mortality in c-avsd patients in univariable logistic regression analyses were surgical era (OR 0.13, 95% CI: 0.04 to 0.40, < p 0.001), preoperative AV valve insufficiency (p = 0.021; moderate AV valve insufficiency: OR 5.09, 95% CI: 1.59 to 16.30, p = 0.006; severe AV valve insufficiency: OR 4.33, 95% CI: 1.09 to 17.22, p = 0.037), and cleft closure, with patients in whom the cleft was closed being at less risk (OR 0.29, 95% CI: 0.11 to 0.78, p = 0.014; Table 3). The multivariable logistic regression analysis revealed surgical era (OR 0.05, 95% CI: 0.01 to 0.19, p < 0.001), age (OR 0.13, 95% CI: 0.02 to 0.83, p = 0.031), and preoperative AV valve insufficiency (p= 0.032; moderate AV valve insufficiency: OR 5.28, 95% CI: 1.50 to 18.60, p = 0.010; severe AV valve insufficiency: OR 4.11, 95% CI: 0.92 to 18.40, p = 0.065) to be independent risk factors. Reoperation Eight patients (2.6% [c-avsd, n = 7; p-avsd, n = 1]) required an early reoperation. Indication for early reoperation was residual ASD/VSD in 3 patients, tricuspid valve insufficiency/stenosis in 2 patients (1 p-avsd patient), left AV valve incompetence in 1 patient, cleft dehiscence in 1 patient, and pulmonary artery stenosis in 1 patient with AVSD- TOF. Three patients (c-avsd, n = 2; p-avsd, n = 1) died shortly after their reoperation (in-hospital deaths and included in the figures above). None of the 5 other patients died during follow-up or required a second reoperation. Forty-three of the 278 hospital survivors (15.5%) who needed no early reoperation required a late reoperation, the median interval being 2.2 years (range, 2 months to 21.1 years) after the primary repair. Ten patients underwent a reoperation within 1 year. Table 4 shows reoperation rates and primary causes for late reoperation for the three AVSD groups. For c-avsd patients (n = 24), the median interval till late reoperation was 2.4 years (range, 3 months to 10.2 years); for p-avsd patients (n = 12), it was 4.0 years (range, 2 months to 21.1 years); and for i-avsd patients (n = 7), it was 2.0 years (range, 1.6 to 13.7). Among the hospital survivors needing a reoperation, there were 4 late deaths, of which 3 were cardiac related (all i-avsd patients). Fourteen patients needed a second reoperation (c-avsd patients, n = 7; p- AVSD patients, n = 5; i-avsd patients, n = 2), mainly for left AV valve incompetence (n = 10). In case of AV valve incompetence, the valve was most frequently repaired at first reoperation, but then replaced at second reoperation (Table 5). 33

13 34

14 Estimated freedom from late reoperation for all hospital survivors without an early reoperation (n = 278) was 96.4% at 1 year, 89.3% at 5 years, and 81.8% at 15 years. Reoperation rate did not significantly differ between c-avsd, p-avsd, and i-avsd patients (Fig 3). For c- AVSD patients, estimated freedom from late reoperation was 96.1% at 1 year, 90.1% at 5 years, and 83.8% at 15 years. For p-avsd patients, it was 95.9% at 1 year, 91.8% at 5 years, and 85.0% at 15 years. For i-avsd patients, it was 100% at 1 year and 75.4% at 5 years. Figures for i-avsd patients at 10 and 15 years could not be reliably estimated. 35

15 Predictors of Late Reoperation In univariable Cox regression analyses for all hospital survivors without an early reoperation (n = 278), risk factors for late reoperation were associated cardiovascular anomalies (hazard ratio [HR] 2.84, 95% CI: 1.51 to 5.33, p = 0.001), left AV valve dysplasia (HR 8.50, 95% CI: 4.59 to 15.73, p < 0.001), preoperative AV valve insufficiency (p = 0.019; moderate AV valve insufficiency: HR 1.21, 95% CI: 0.54 to 2.75, p = 0.645; severe AV valve insufficiency: HR 2.59, 95% CI: 1.24 to 5.41, p = 0.011), and cleft closure (HR 0.39, 95% CI: 0.21 to 0.72, p = 0.002; Table 6). The multivariable Cox regression analysis showed associated cardiovascular anomalies (HR 4.04, 95% CI: 2.08 to 7.84, p < 0.001), left AV valve dysplasia (HR 8.26, 95% CI: 4.38 to 15.58, p < 0.001), and cleft closure (HR 0.39, 95% CI: 0.21 to 0.73, p = 0.003) to be independent risk factors. To assess whether the risk for patients with TOF differed from that for patients with other associated cardiovascular anomalies, a second multivariable Cox regression analysis was performed with associated cardiovascular anomalies being redefined as absent, TOF, or other. In this analysis, both subgroups were at higher risk for late reoperation compared with patients who did not have any associated cardiovascular anomaly (p < 0.001; TOF: HR 5.05, 95% CI: 1.94 to 13.14, p _ 0.001; other associated cardiovascular anomalies: HR 3.63, 95% CI: 1.68 to 7.84, p = 0.001). The HRs for left AV valve dysplasia and cleft closure became 8.57 (95% CI: 4.47 to 16.44, p < 0.001) and 0.38 (95% CI: 0.20 to 0.72, p < 0.001), respectively. Separate analyses were performed to identify risk factors for late reoperation for the three AVSD groups. For c-avsd patients, the univariable analyses revealed surgical era (HR 0.31, 95% CI: 0.13 to 0.74, p = 0.009), associated cardiovascular anomalies (HR 2.64, 95% CI: 1.15 to 6.05, p = 0.022), left AV valve dysplasia (HR 13.37, 95% CI: 5.81 to 30.80, p < 0.001), preoperative AV valve insufficiency (p < 0.001; moderate AV valve insufficiency: HR 1.75, 95% CI: 0.57 to 5.43, p = 0.332; severe AV valve insufficiency: HR 10.20, 95% CI: 3.83 to 27.20, p < 0.001), and cleft closure (HR 0.18, 95% CI: 0.08 to 0.40, p < 0.001) to be risk factors (Table 6). In the multivariable analysis, left AV valve dysplasia (HR 4.96, 95% CI: 1.80 to 13.72, p = 0.002), preoperative AV valve insufficiency (p = 0.011; moderate AV valve insufficiency: HR 1.32, 95% CI: 0.41 to 4.21, p = 0.644; severe AV valve insufficiency: HR 5.06, 95% CI: 1.57 to 16.36, p = 0.007), and cleft closure (HR 0.16, 95% CI: 0.07 to 0.37, p, 0.001) remained as independent risk factors. In the multivariable analysis with the variable associated cardiovascular anomalies being redefined, associated cardiovascular anomalies was also an independent risk factor with only 36

16 37

17 only patients with TOF being at higher risk (p = 0.003; TOF: HR 6.98, 95% CI: 2.27 to 21.51, p = 0.001; other cardiovascular anomalies: HR 0.94, 95% CI: 0.24 to 3.62, p = 0.925). In this analysis, HRs for the other variables became 5.34 for left AV valve dysplasia (95% CI: 1.73 to 16.49, p = 0.004), 6.13 for severe AV valve insufficiency (95% CI: 1.79 to 21.03, p = 0.004), and 0.13 for cleft closure (95% CI: 0.05 to 0.31, p < 0.001). For p-avsd patients, left AV valve dysplasia was the only risk factor in the univariable analyses (HR 5.76, 95% CI: 1.72 to 19.28, p = 0.004; Table 6). The multivariable analysis showed left AV valve dysplasia (HR 6.74, 95% CI: 1.95 to 23.37, p = 0.003) to be an independent risk factor, and, in addition, associated cardiovascular anomalies (HR 4.51, 95% CI: 1.14 to 17.91, p = 0.032). For i-avsd patients, associated cardiovascular anomalies (HR 8.12, 95% CI: 1.60 to 41.29, p = 0.012) was the only risk factor in the univariable analyses (Table 6), and remained as the only predictor in the multivariable analysis (stepwise method). Comment In this retrospective study, the outcome of primary correction of AVSD in 312 patients who were operated on over a period of 30 years was analyzed. In-hospital mortality for the total study population was 8.3%. This mortality is lower than the rates of 16% and 14.9% that have been reported by Studer and colleagues [9] and Boening and associates [10], respectively, and is comparable to the rates between 5.7% and 10% reported by others [2 4, 11]. Not all studies did focus on the same study population as we did. Some reports focused on c-avsd only [2 4]. In-hospital mortality for our c- AVSD patient group was 10.5%. We found much lower early mortality rates for our p-avsd patients (3.9%) and i-avsd patients (3.7%). Yet, type of AVSD appeared not to be a risk factor for early mortality in our logistic regression analyses. Many reports describe improved surgical results after AVSD repair over the last decade, although recent figures for in-hospital mortality after primary repair still vary between 2.5% and 10% [1, 2, 9, 12]. We found early surgical era to be an independent risk factor for inhospital mortality. In our series, in-hospital mortality decreased from 13.3% for the period to 1996, to 3.2% since Among our c-avsd patients, it decreased from 20.7% to 3.3%. Reasons for improved survival in the more recent era include improved accuracy of preoperative diagnosis, refined surgical techniques and advances in intraoperative support (eg, the introduction of intraoperative transesophageal echocardiography), and better postoperative 38

18 management. The timing of the operation is an important issue. Correction of the AVSD should be performed before development of irreversible pulmonary vascular changes [10, 13]. Other important arguments for early repair are a possible increase in degenerative changes of the AV valve and progressive ventricular dilation with aging [1, 10, 12]. Moreover, early repair avoids the use of palliative procedures. As in other studies [14], there was an increase in the rate of repair of AVSD during infancy in our series. Our current policy is to operate upon c- AVSD and i-avsd in the first 3 months of life; p-avsd is repaired in the first 2 years. Together with the increase in operations at younger age, in-hospital mortality in our study group of infants younger than 6 months decreased from 37.9% before 1996 to 4.4% since Increase in operations at a younger age with low early mortality has also been reported by others [15]. This observation seems to conflict with the fact that we found younger age to be an independent risk factor for early mortality. However, although in the earlier decades of our experience, young age was definitively a risk factor for mortality, we now see that even with the operative age being much lower than it was ever before, results have impressively improved. In our study group, preoperative AV valve insufficiency showed a tendency toward statistical significance as an independent risk factor for early mortality. This observation is in contrast to the findings that were reported by Tweddell and colleagues [14], but is consistent with findings of other groups [9, 12]. We believe that the pathophysiologic changes over time induced by moderate to severe AV valve insufficiency, including annular dilation, stiffening and curling-in of the leaflets, as well as lengthening of tendinous chords, justify early correction. We found no statistically significant difference in overall survival between c-avsd, p- AVSD, and i-avsd patients. Long-term overall survival of c-avsd patients is reported by others to be 88% at 5 years [9], 82% at 10 years [11], and 76% at 15 years [16]. Estimated overall survival in our population of c-avsd was a little higher in the long-term, being 88.3% at 5 and 15 years. Regarding the estimated overall survival of i-avsd and p-avsd patients, our results are comparable with those of other authors [5]. Late reoperation was performed in 15.5% of our hospital survivors who needed no early reoperation. This is comparable with the figures found by other authors [10, 17]. The main indication for late reoperation was left AV valve regurgitation, which could be repaired at reoperation in the majority of the patients. Late reoperation rate was found to be not significantly different between the different types of AVSD. Multivariable Cox regression 39

19 analysis showed associated cardiovascular anomalies, left AV valve dysplasia, and absence of cleft closure to be independent risk factors for late reoperation. We, therefore, strongly advise closing the cleft during repair of AVSD, as do others [5, 18]. Study Limitations Some limitations to this study must be recognized. First, this is a retrospective, observational study. Although all patients had clear documentation of preoperative variables such as left AV valve dysplasia and left AV valve regurgitation, other interesting variables such as pulmonary hypertension and elevated pulmonary vascular resistance were less well documented and could not be examined in relation to outcome. Second, as the study covered more than 30 years of experience, a learning curve not only regarding the procedure itself but also for patient selection might have influenced results in the early days. In addition, anesthetic and cardiopulmonary bypass techniques, intraoperative support, and postoperative management evolved over time. To account for the time factor, surgical era was used as a possible predictor of outcome in all regression analyses. Two surgical eras were defined mainly based on a shift toward a younger patients age at repair from 1996 (Fig 1). Finally, patient numbers for p-avsd and i-avsd subgroups were relatively small, limiting the analyses with regard to outcome for these subgroups. Also, the numbers for the different types of associated cardiovascular anomalies were small and varied over subgroups. Consequently, potential risks related to specific anomalies were difficult to assess. In conclusion, AVSD repair can be accomplished with good long-term results. Risk factors for early mortality were surgical era before 1996 and younger age at surgery. Risk factors for late reoperation were associated cardiovascular anomalies, left AV valve dysplasia, and absence of cleft closure. As this series comprises the experience of more than 30 years, the study results need to be regarded in this perspective. It should be noted that there was a strong decline in age at AVSD repair in the last decade, and at the same time, a significant decrease in in-hospital mortality. 40

20 References 1. Yasui H, Nakamura Y, Kado H, et al. Primary repair for complete atrioventricular canal: recommendation for early primary repair. J Cardiovasc Surg 1990;31: Alexi-Meskishvili V, Ishino K, Dähnert I, et al. Correction of complete atrioventricular septal defects with the double-patch technique and cleft closure. Ann Thorac Surg 1996;62: Bando K, Turrentine MW, Sun K, et al. Surgical management of complete atrioventricular septal defects. J Thorac Cardiovasc Surg 1995;110: Hanley FL, Fenton KN, Jonas RA, et al. Surgical repair of complete atrioventricular canal defects in infancy. Twentyyear trends. J Thorac Cardiovasc Surg 1993;106: El-Najdawi EK, Driscoll DJ, Puga FJ, et al. Operation for partial atrioventricular septal defect: a forty-year review. J Thorac Cardiovasc Surg 2000;119: Weintraub RG, Brawn WJ, Venables AW, Mee RB. Two-patch repair of complete atrioventricular septal defect in the first year of life. Results and sequential assessment of atrioventricular valve function. J Thorac Cardiovasc Surg 1990;99: Hoohenkerk GJ, Schoof PH, Bruggemans EF, Rijlaarsdam M, Hazekamp MG. 28 Years experience with transatrialtranspulmonary repair of atrioventricular septal defect with tetralogy of Fallot. Ann Thorac Surg 2008;85: Hoohenkerk GJ, Wenink AC, Schoof PH, et al. Results of surgical repair of atrioventricular septal defect with doubleorifice left atrioventricular valve. J Thorac Cardiovasc Surg 2009;138: Studer M, Blackstone EH, Kirklin JW, et al. Determinants of early and late results of repair of atrioventricular septal (canal) defects. J Thorac Cardiovasc Surg 1982;84:

21 10. Boening A, Scheewe J, Heine K, et al. Long-term results after surgical correction of atrioventricular septal defects. Eur J Cardiothorac Surg 2002;22: McGrath LB, Gonzalez-Lavin G. Actuarial survival, freedom from reoperation, and other events after repair of atrioventricular septal defects. J Thorac Cardiovasc Surg 1987;94: Pozzi M, Remig J, Fimmers R, Urban AE. Atrioventricular septal defects. Analysis of short- and medium-term results. J Thorac Cardiovasc Surg 1991;101: Newfeld EA, Sher M, Paul MH, Nikaidoh H. Pulmonary vascular disease in complete atrioventricular canal defect. Am J Cardiol 1977;39: Tweddell JS, Litwin SB, Berger S, et al. Twenty-year experience with repair of complete atrioventricular septal defects. Ann Thorac Surg 1996;62: Kortenhorst MS, Hazekamp MG, Rammeloo LA, Schoof PH, Ottenkamp J. Complete atrioventricular septal defect in children with Down syndrome: good results of surgical correction at younger age. Ned Tijdschr Geneeskd 2005;149: Michielon G, Stellin G, Rizzoli G, et al. Left atrioventricular valve incompetence after repair of common atrioventricular canal defects. Ann Thorac Surg 1995;60:S Gunther T, Mazzitelli D, Haehnel CJ, Holper K, Sebening F, Meisner H. Long-term results after repair of complete atrioventricular septal defects: analysis of risk factors. Ann Thorac Surg 1998;65: Wetter J, Sinzobahamvya N, Blaschczok C, et al. Closure of the zone of apposition at correction of complete atrioventricular septal defect improves outcome. Eur J Cardiothorac Surg 2000;17:

22 INVITED COMMENTARY Surgical outcomes for congenital heart disease have greatly improved during the last several decades. These improvements are associated with enhancements in surgical and anesthetic techniques, cardiopulmonary bypass and cardiac protection, and postoperative care strategies. However, contemporary outcomes can not be measuredalone by mortality rates, which have decreased dramatically, but must also be centered on lessening morbidity. Decreasing morbidity would thus primarily lead to a better patient and family experience, and it would also beget less resource use. Hoohenkerk and colleagues [1] present a 30-year experience with atrioventricular septal defect (AVSD) repair. The investigators evaluated risk factors associated with mortality and reoperation. As one might expect, the surgical era had a positive effect on mortality. The era effect exemplifies the multifactorial nature of improved outcomes, astutely noting that earlier age at operation for those at risk for irreversible pulmonary vascular disease played a role. However, certain factors were associated with late reoperation; these included other cardiac anomalies, left atrioventricular valve dysplasia, and absence of cleft closure at the initial operation. This observation is congruent with other experiences noted in the literature. Long-term studies such as this help determine expected outcomes, improve anticipatory guidance to families, and further allow the analysis of risk factors that predict both longer hospital stays and complications versus unplanned interventions. In addition, these studies also lead to future thoughts to ponder. In short, the outcomes presented are commensurate with expected mortality in the current era. Further study may be necessary. 43

23 44