Surgical repair of partial atrioventricular septal defects (AVSD) has been successful for more than 60 years. However, recent data from the Pediatric Heart Network show that 31% of patients have moderate or severe left atrioventricular valve regurgitation (LAVVR) at follow-up. Previously, our institution found that only 9% of patients had more than moderate LAVVR at the last follow-up. Our objective was to determine the long-term outcomes after repair of partial AVSD in the current era. We reviewed all patients with partial AVSD who had primary biventricular repair from January 1995 to June 2011 at our institution. The Kaplan-Meier method was used to estimate the survival free of an event, and factors were evaluated for an association with each outcome using the log-rank test. All 105 patients with partial AVSD who had surgery during this period were evaluated. The median age at surgery was 7.9 years. The overall survival rate at 1 year was 97%. Median follow-up was 5.3 years (interquartile range 1.7 to 11.1). At 3 years, the survival rate free from reoperation was 89%. Thirteen patients required reoperations with the most common reason being LAVVR. A total of 10 patients developed more than moderate LAVVR with a cumulative incidence of 8% by 2 years. The discrepancy with the Pediatric Heart Network data may be due to the later age of operation for patients in our cohort suggesting that elective repair of partial AVSD should be deferred until children are somewhat older (ages 5 to 8 years). Neither patient age (p = 0.11) nor severity of preoperative LAVVR (p = 0.16) were identified as statistically significant risk factors. In conclusion, there is less morbidity and mortality after surgical repair for partial AVSD.
Surgical repair of partial atrioventricular septal defects (AVSD) has been successful for more than 60 years. Many studies have reported good early and late outcomes, but recent data from the Pediatric Heart Network (PHN) show that 31% of patients had moderate or severe left atrioventricular valve regurgitation (LAVVR) at 6-month follow-up. Previous reports have found a lower rate of important LAVVR at follow-up. An earlier study from our institution found that only 9% of patients developed more than moderate LAVVR. The purpose of this study was to describe the long-term outcome after surgical repair of partial AVSD in the current era and compare those results with the PHN data (2004 to 2006) and the earlier Mayo Clinic experience (1955 to 1995). We estimated the overall survival rate, rate of survival free of reoperation, and cumulative incidence of significant LAVVR, left atrioventricular valve (LAVV) stenosis, left ventricular outflow tract obstruction (LVOTO), and arrhythmia in our cohort. In addition, we administered the Pediatric Quality of Life Inventory to determine the health-related quality of life (QOL) in patients after repair of partial AVSD.
Methods
In this institutional review board–approved study, we performed a retrospective chart review of all patients who had primary repair of partial AVSD or transitional (primum atrial septal defect [ASD], cleft LAVV, and a restrictive ventricular septal defect ) AVSD from January 1995 to June 2011 at Mayo Clinic, Rochester, Minnesota. Patients with common atrium, heterotaxy syndromes, and right ventricular outflow obstruction were excluded from the study.
Data collected from each patient’s record included demographics, anatomic details of the cardiac defect and associated anomalies, and operative data. Serial echocardiographic data were collected from preoperative, hospital dismissal, 6-month or first postoperative visit, and the most recent visit if available. Passive follow-up information was obtained from clinical charts to establish vital status and outcome events of interest. Active follow-up was obtained with a detailed health status questionnaire sent to each known survivor. Patients who did not return or did not complete the questionnaire were sent a second questionnaire. If the second questionnaire was not returned, an attempt was made to contact the patients by telephone. Data collected from the mailed questionnaire included current health status and symptoms, medications, arrhythmia history, subsequent hospitalizations, and functional status.
In addition, each patient received a Pediatric Quality of Life Inventory. This inventory consisted of the 23-item PedsQL 4.0 Generic Core Scales that encompassed 4 domains: physical functioning, emotional functioning, social functioning, and school functioning. The PedsQL scales are composed of parallel child self-report and parent proxy report formats. Items are linearly transformed to a 0-to-100 scale, so that higher scales indicate better QOL. In addition, there are 3 summary scores: physical health summary score, psychosocial health summary score, and total scale score. The reliability and validity of the PedsQL Generic Core Scales have been demonstrated in healthy and patient populations.
LAVVR severity was based on the severity recorded in the medical record or echocardiographic report and considered to be important if graded as moderate or greater. Regurgitation was graded subjectively based on the color Doppler jets and chamber size. LAVV stenosis was considered important if the mean gradient was >6 mm Hg. LVOTO was considered important if the patient had reoperation for relief of obstruction or if there was a left ventricular outflow maximum instantaneous gradient ≥20 mm Hg measured with continuous-wave Doppler.
Results are summarized using standard descriptive statistics. Duration of follow-up was calculated from the date of the AVSD repair to the date of last follow-up or death. The overall survival rate was estimated using the Kaplan-Meier method. In addition, the Kaplan-Meier method was used to estimate the survival rate free of a reoperation. For outcomes assessed with echocardiography, such as LAVVR, duration of follow-up was calculated from the date of the AVSD repair to the date of first echocardiography documenting moderate or severe LAVVR. Otherwise, follow-up was censored at the date of the last echocardiography. Factors were evaluated univariately for their association with the time-to-event outcomes using the log-rank test. The gender and age at surgery were compared between the questionnaire responders and nonresponders using the chi-square test and Wilcoxon rank-sum test, respectively. All calculated p values were 2-sided, and p values <0.05 were considered statistically significant. Analyses were performed using SAS, version 9.2, statistical software (SAS Institute, Inc, Cary, North Carolina).
Results
The study group consisted of 105 patients ( Table 1 ; 84 with partial AVSD and 21 with transitional AVSD). Nineteen patients (18%) had Down syndrome. The median age at operation was 7.9 years (interquartile range [IQR] 1.5 to 27.2, range 11 days to 69.9 years). The most common associated defect was a secundum ASD, which was observed in 25 patients (24%; Table 2 ). Double-orifice LAVV was present in 3 patients, and a parachute LAVV was present in 5 patients.
| Characteristic | |
|---|---|
| Gender, n (%) | |
| Male | 42 (40%) |
| Female | 63 (60%) |
| Down syndrome, n (%) | 19 (18%) |
| Age at surgery (years) | |
| Median (range) | 7.9 (0.03–69.9) |
| <2 | 30 (29%) |
| 2–9 | 25 (24%) |
| 10–18 | 14 (13%) |
| 19–29 | 11 (10%) |
| 30 and older | 25 (24%) |
| Type of atrioventricular septal defect (AVSD) | |
| Partial | 84 (80%) |
| Transitional | 21 (20%) |
| Associated Defect ∗ | N (% of 105) |
|---|---|
| Secundum atrial septal defect | 25 (24%) |
| Patent ductus arteriosus | 9 (9%) |
| Left superior vena cava | 6 (6%) |
| Parachute left atrioventricular valve | 5 (5%) |
| Abnormal systemic veins | 4 (4%) |
| Abnormal pulmonary veins | 4 (4%) |
| Double orifice left atrioventricular valve | 3 (3%) |
| Pulmonary valve stenosis | 2 (2%) |
| Coarctation of the aorta | 2 (2%) |
| Left ventricular outflow tract obstruction | 2 (2%) |
| Ebstein anomaly | 2 (2%) |
| Other † | 4 (4%) |
∗ Individual patients may have multiple associated defects.
† Right ventricular outflow tract obstruction, transposition of the great arteries, cor triatriatum dexter, and dilation of the ascending aorta.
Ninety-eight patients (94%) had a single-patch closure of the primum ASD, whereas 4 patients (4%) had suture closure and 1 patient with a transitional defect had double-patch closure of the ASD and a restrictive ventricular septal defect. The cleft in the LAVV was closed in 100 patients (95%), and LAVV posterior annuloplasty was performed in 20 patients (19%). Other cardiac operations included suture closure of a ventricular septal defect (8), right atrioventricular valve annuloplasty (14), right atrioventricular valve repair (2), right atrioventricular valve replacement (2), resection of LVOTO (2), maze procedure (4), repair of anomalous pulmonary veins (2), right reduction atrioplasty (3), patent ductus arteriosus ligation (2), left superior vena cava ligation (1; we excluded patients with right ventricular outflow tract obstruction), reduction of ascending aorta (1), and arterial switch (1). Median hospital stay was 5 days (IQR 4 to 6). The median cardiopulmonary bypass time was 63 minutes (IQR 50 to 81).
Early mortality was <1% (1 of 105); this patient died in the hospital after a stroke. An additional patient died suddenly, <30 days after the AVSD repair, at home from an unknown cause. There was 1 additional death within the first postoperative year due to sepsis, yielding an overall survival rate at 1 year of 97% (95% confidence interval [CI] 93.3% to 100%). All 3 of the patients who died within 1 year of surgery were infants <1 year of age at the time of the surgery. Two late deaths have been documented, but neither was primarily related to the surgical repair. The overall survival rate at 5 years was 95% (95% CI 90.4% to 100%; Figure 1 ). Median follow-up of the remaining 100 patients was 5.3 years (IQR 1.7 to 11.1). Patients with Down syndrome had a similar survival rate compared with the rest of the cohort (p = 0.88, 5-year survival rate: 94% vs 95%, respectively). On dividing the patients into quartiles based on their age at surgery, the data suggest that patients operated at a very young age tend to have worse survival (p = 0.045). Among patients aged <1.5 years at the time of surgery, the overall survival rate at 5 years was 88%. This compares with the 100% survival rate for those aged 1.5 to 8 years, 93% for those aged 8 to 27 years, and 100% for those aged >27 years at the time of surgery.
Thirteen patients required reoperation. The primary reason for reoperation included LAVVR (n = 7), LAVV stenosis (n = 1), LAVVR and residual ASD (n = 1), relief of LVOTO (n = 3), and LAVVR and relief of LVOTO (n = 1). The median interval between the primary repair and the subsequent reoperation for these 13 patients was 2 years (range 8 days to 13.3 years). The rate of survival free of reoperation was 96% at 1 year (95% CI 91.2% to 99.0%), 89% at 3 years (95% CI 81.8% to 96.0%), and 87% at 5 years (95% CI 79.7% to 95.0%; Figure 2 ).
All patients had a documented postoperative echocardiography. The median duration of time from the date of repair until the most recent echocardiography was 1.1 years (IQR 6 days to 4.2 years). Data were abstracted regarding the patients’ preoperative, dismissal, 6-month, and most recent echocardiographic examination. Of the 102 patients who survived past the first postoperative year, 66 had their last echocardiogram read at our institution <3 years after surgery, 11 had their last echocardiogram from 3 to 5 years after surgery, and the remaining 25 patients had their last echocardiogram >5 years after surgery.
Ten patients developed more than moderate LAVVR within 10 years after surgery with a cumulative incidence of 3% (95% CI 0% to 6.9%) by 6 months, 6% (95% CI 1.0% to 11.5%) by 1 year, 10% (95% CI 2.0% to 17.4%) by 3 years, and 13% (95% CI 3.1% to 21.6%) by 5 years postoperative. None of the patients with Down syndrome developed moderate or severe LAVVR (p = 0.13), and gender (p = 0.32), preoperative LAVVR (none, trivial, or mild vs moderate or severe; p = 0.16), or age quartiles (p = 0.33) were not identified as statistically significant risk factors. Among patients aged <1.5 years at the time of surgery, the cumulative incidence by 3 years was 15% (95% CI 0% to 29%), compared with 7% (95% CI 0% to 18%), 15% (95% CI 0% to 32%), and 0% for patients 1.5 to 8 years, 8 to 27 years, and >27 years, respectively, at the time of surgery. Of the 9 patients who had reoperation for LAVVR, 6 patients had LAVV replacement and 3 had valve repair.
Two patients developed LAVV stenosis after operation. One occurred 8 days postoperative and required removal of an annuloplasty stitch. The second developed severe LAVV stenosis and regurgitation and required valve replacement.
Six patients developed LVOTO with a cumulative incidence of 3% (95% CI 0% to 7.0%), 9% (95% CI 1% to 15.8%), and 11% (95% CI 2% to 20.5%) by 1, 3, and 5 years, respectively, after the initial surgery. Four patients required reoperation to relieve LVOTO.
Six patients had a residual ASD. One patient had closure of the ASD during reoperation for LAVVR.
Eight patients had postoperative arrhythmias. These arrhythmias included supraventricular tachycardia (2), atrial flutter (4), and 2 patients who noted a catheter ablation procedure on their follow-up questionnaire, but their underlying arrhythmia was unknown. Three patients in total required postoperative radiofrequency ablation for supraventricular tachycardia or atrial flutter. No pacemakers were implanted.
Of the 100 patients known to be alive, 51 responded to the questionnaire at a median of 8.6 years (IQR 5.3 to 14.3, range 1.3 to 16.7) after their initial operation. The median patient age at the time of the questionnaire was 26.5 years (IQR 14.1 to 38.2, range 2.3 to 75.9). During the follow-up period, no patients developed bacterial endocarditis. There was no significant difference in the gender of the responders and nonresponders (37% vs 43%, respectively, were men, p = 0.58). The responders were older than the nonresponders at the time of the surgery; however, the difference did not reach statistical significance (median 14.2 vs 5.0 years, p = 0.26).
Parent proxy report responses to the PedsQL Inventory were available for 23 patients aged <19 years at the questionnaire. Results are presented for the 20 patients without Down syndrome. Four (20%) of the 23 patients had at least 1 item scored as significant (“often” or “almost always” a problem) by their parent proxy. As listed in Table 3 , 15% of the parent proxies reported a significant problem related to physical functioning, 10% related to emotional functioning, 10% related to social functioning, and 16% related to school functioning problems with 4 parent proxies (20%) reporting patients having significant problems in >1 domain. Patients were classified as “at risk for impaired QOL” if their domain score was >1 SD below the population mean for healthy subjects. Among these 20 patients, the percentage of patients classified as at risk for impaired QOL for each domain is listed in Table 3 .
