Abstract
Objectives
We sought to identify factors associated with timing and surgical outcomes of the superior cavopulmonary anastomosis (SCPA).
Methods
The Pediatric Heart Network’s Infant Single Ventricle trial database identified participants who underwent SCPA. Factors potentially associated with age at SCPA, length of stay (LOS) and death by 14 months of age were evaluated. Factors included subject demographics, cardiac anatomy, measures from neonatal hospitalization and pre-SCPA visit, adverse events, echocardiographic variables, intraoperative variables, SCPA type, and the number of concurrent cardiac surgical procedures. Age at SCPA was analyzed by Cox proportional hazards regression. Natural log LOS was analyzed by multiple linear regression.
Results
SCPA was performed in 193 subjects at 5.2 (IQR 4.2, 6.2) months of age and weight of 5.9 (IQR 5.3, 6.6) kg. Median LOS was 7 days (IQR 6, 10). There were 3 deaths and 1 transplant during SCPA hospitalization and 3 deaths and 3 transplants between discharge and 14 months of age. Age at SCPA was associated with center and interstage adverse events. Longer LOS was associated with younger age and greater case complexity. SCPA type, valve regurgitation, ventricular ejection fraction and ventricular end-diastolic pressure were not independently associated with age at SCPA or LOS.
Conclusions
Greater case complexity and more frequent interstage adverse events are associated with earlier age at SCPA. Significant variation in age at SCPA among centers, independent of subject factors, highlights a lack of consensus regarding optimal timing. Factors associated with LOS may offer insights for improving pre-SCPA care and surgical outcome.
Keywords: Heart defects, congenital, single ventricle
Background
The vast majority of children with single ventricle physiology undergo superior cavopulmonary anastomosis (SCPA) as part of staged palliation. Factors associated with surgical outcomes have been addressed predominantly by single center studies with varying and often conflicting results. These studies have identified age, weight-forage, pulmonary artery pressure, ventricular morphology, atrioventricular valve regurgitation, and cardiopulmonary bypass time as variables associated with SCPA outcomes.(1–7) However, optimal timing of the procedure remains a topic of debate and varies by center, in part, due to lack of consistent associations across these studies. This practice variation across centers may also play a role in determining outcomes.(8–10)
With detailed clinical data from a multicenter network, the Infant Single Ventricle (ISV) trial database offered a unique opportunity to evaluate factors associated with the timing of SCPA and subsequent surgical outcomes in infants undergoing SCPA. This study sought to identify factors associated with SCPA timing and surgical outcomes.
Methods
From August 2003 through July 2008, the Pediatric Heart Network conducted a randomized, double-blind, placebo-controlled clinical trial comparing the effects of angiotensin-converting enzyme (ACE) inhibitor with those of placebo on somatic growth in infants with single ventricle (ClinicalTrials.gov Identifier NCT00113087). Infants ≤45 days of age with single ventricle physiology, stable systemic and pulmonary blood flow, and planned SCPA were enrolled at 10 North American centers. Subjects were followed from enrollment until 14 months of age. No significant differences were found between placebo and treatment groups. Detailed descriptions of the ISV trial design and methods, including enrollment, drug allocation, follow-up, data analysis, and results have been published.(11, 12)
The trial database was queried for all infants undergoing SCPA during the study period. Age at SCPA was the outcome for surgical timing. Death, heart transplantation, hospital length of stay, unanticipated surgical or cardiac catheterization interventions following SCPA, days on ventilator, and days until last chest tube removal were clinical outcomes of interest. Unanticipated interventions included all post-SCPA cardiac catheter interventions and additional cardiac surgeries that occurred before discharge, excluding chest closure. The number of unanticipated interventions was dichotomized as none or ≥1; total days on ventilator as ≤2 days or >2 days; and days until last chest tube removal as ≤5 days or >5 days post-SCPA. These thresholds for total days on ventilator and days until last chest tube represent a consensus opinion of suboptimal outcomes by the authors but also approximated the highest quartile of days (rounded to an even number). Subjects who died or had heart transplantation during the SCPA hospitalization were excluded from the analyses of SCPA hospital length of stay, days on ventilator, and days until last chest tube removal. Potentially associated factors were selected from subject demographics, cardiac anatomy, measures from the neonatal hospitalization and the clinical assessments at the pre-SCPA visit, adverse events during the interstage time period between discharge from neonatal hospitalization and SCPA, type of SCPA performed, and the number of concurrent cardiac surgical procedures at SCPA (APPENDIX 1). Adverse events were classified as non-serious, moderately serious and serious using the Common Terminology Criteria for Adverse Event v.3.0.(13) All serious adverse events were adjudicated by an independent physician who was unaware of treatment group assignment. Ventricular mass and volume measured by a core laboratory, and mass:volume ratio were expressed as z-scores relative to body surface area and age respectively.(14) Status of ACE inhibitor use was determined by whether the subject was taking any form of ACE inhibitor (either assigned study drug or open label medication) at the time of the pre-SCPA clinical visit. Brain natriuretic peptide (BNP) was obtained at the time of the pre-SCPA clinic visit and was measured at a central laboratory.
Statistical analysis
Descriptive statistics for subject and SCPA hospitalization characteristics are presented as median and inter-quartile range for continuous measures with a skewed distribution, mean and standard deviation for other continuous measures, or frequency and percentage for categorical variables.
Age at SCPA was analyzed using Cox proportional hazards regression which allowed for the introduction of inter-stage adverse events as time-varying covariates. Length of stay (LOS) was analyzed after natural log transformation using multivariable linear regression. The linearity of associations with LOS was explored using generalized additive models. Dichotomous clinical outcomes were analyzed using logistic regression with Firth’s penalized likelihood applied when center was included in the models. To identify any possible nonlinear associations with the dichotomous outcome and time-to-event outcome, tertiles, median cut-offs, as well as square root or natural log transformations of potential associated factors were evaluated in logistic regression and Cox proportional hazards modeling.
Variables were considered in the multivariable model selection when the univariate or center-adjusted association was significant at the level p≤0.15. All models were adjusted for center to account for any differences in clinical management strategies between centers that were not fully captured by the ISV database. Stepwise selection methods were used to determine all final models; predictors with p ≤ 0.05 remained in the model. The effects of mean pulmonary artery pressure (N=98) and transpulmonary gradient (N=60) were explored in univariate and multivariable analysis, but were not included in the final modeling due to the large amount of missing data. Missing values were not imputed. Statistical analyses were performed using SAS Statistical Software v.9.2 (SAS Institute, Cary NC).
Results
Subject and SCPA hospitalization characteristics
Of the 230 subjects randomized in the trial, 193 (84%) underwent SCPA. There were 5 randomized subjects who did not undergo SCPA but remained in the trial. The remainder withdrew from the trial prior to SCPA for the following reasons: death (18), family preference (8), heart transplantation (3), physician preference (1), lost to follow-up (1), and other (1).
Tables 1 and 2 describe the details of the SCPA hospitalization overall and by type of SCPA. With the exception of the proportion of subjects having aortopulmonary shunt takedown, subject characteristics did not significantly differ by the type of SCPA performed. There was a trend (p = 0.06) toward more arrhythmia in subjects undergoing Hemi-Fontan procedures.
Table 1.
Summary of the Characteristics of the SCPA Hospitalization
| All Subjects (N=193) | Type of SCPA
|
||||||
|---|---|---|---|---|---|---|---|
| BBDCPA1 (N=28) | BDCPA1 (N=134) | Hemi-Fontan (N=23) | Hemi-Fontan, with BCA1 (N=3) | Kawashima (N=5) | P-value2 | ||
| Age at SCPA (months) | |||||||
| Median (IQR) | 5.2 (4.2, 6.2) | 5.4 (4.6, 6.9) | 5.1 (4.1, 6.0) | 5.6 (4.1, 6.0) | 5.0 (4.6, 7.1) | 6.7 (5.3, 7.5) | 0.36 |
| Gender | 0.89 | ||||||
| Male | 136 (70) | 20 (71) | 94 (70) | 16 (70) | 3 (100) | 3 (60) | |
| Female | 57 (30) | 8 (29) | 40 (30) | 7 (30) | 0 | 2 (40) | |
| Race | 0.34 | ||||||
| White | 153 (80) | 19 (68) | 110 (83) | 18 (78) | 3 (100) | 3 (60) | |
| Black | 27 (14) | 7 (25) | 17 (13) | 2 (9) | 0 | 1 (20) | |
| Other | 12 (6) | 2 (7) | 6 (5) | 3 (13) | 0 | 1 (20) | |
| Ventricular Morphology | 0.027 | ||||||
| Left | 34 (18) | 3 (11) | 26 (19) | 5 (22) | 0 | 0 | |
| Right | 138 (72) | 17 (61) | 97 (72) | 18 (78) | 3 (100) | 3 (60) | |
| Mixed | 21 (11) | 8 (29) | 11 (8) | 0 | 0 | 2 (40) | |
| Number of Concurrent Cardiac Surgical Procedures | |||||||
| Median (IQR) | 1.0 (1, 2) | 2.0 (1, 2) | 1.0 (1, 2) | 1.0 (0, 2) | 1.0 (1, 2) | 1.0 (0, 1) | 0.50 |
| Surgical Procedures, n(%)3 | |||||||
| Atrial septectomy | 20 (10) | 3 (11) | 12 (9) | 5 (22) | 0 | 0 | 0.43 |
| Patch repair of pulmonary artery stenosis | 71 (37) | 9 (32) | 47 (35) | 11 (48) | 1 (33) | 3 (60) | 0.55 |
| Takedown of aortopulmonary shunt | 30 (16) | 10 (36) | 18 (13) | 1 (4) | 1 (33) | 0 | 0.015 |
| Takedown of RV-PA conduit | 45 (23) | 5 (18) | 34 (25) | 3 (13) | 2 (67) | 1 (20) | 0.25 |
| Hospital Length of Stay (days) | |||||||
| Median (IQR) | 7 (6, 10) | 8.5 (6, 13) | 7.0 (6, 9) | 8.0 (7, 11) | 8.0 (6, 12) | 8 (7, 8) | 0.12 |
| ICU length of stay (days) | |||||||
| Median (IQR) | 4.0 (3, 5) | 5.0 (3, 7) | 4.0 (3, 5) | 4.0 (3, 5) | 3.0 (2, 8) | 4.0 (4, 6) | 0.22 |
| Total Days of Ventilator Support | |||||||
| Median (IQR) | 2.0 (1, 2) | 2.0 (1, 3) | 2.0 (1, 2) | 2.0 (1, 2) | 2.0 (1, 2) | 2.0 (1, 2) | 0.79 |
| Total Days until Last Chest Tube Removed | |||||||
| Median (IQR) | 3.0 (2, 4) | 4.0 (2, 5) | 3.0 (2, 4) | 4.0 (3, 4) | 5.0 (5, 5) | 4.0 (1, 4) | 0.09 |
| Subjects with Post-op Arrhythmia requiring intervention, n (%) | 20 (10) | 2 (7) | 11 (8) | 6 (26) | 1 (33) | 0 | 0.06 |
| Bradycardia requiring artificial pacemaker4, n (%) | 13 (62) | 2 (100) | 7 (58) | 4 (67) | 0 | 0 | 0.57 |
| Other Arrhythmia4, n (%) | 4 (19) | 0 | 3 (25) | 0 | 1 (100) | 0 | 0.16 |
| Number of interventional cardiac catheterizations, n (%) | 1.00 | ||||||
| 0 | 184 (95) | 27 (96) | 126 (94) | 23 (100) | 3 (100) | 5 (100) | |
| 1 | 4 (2) | 0 | 4 (3) | 0 | 0 | 0 | |
| ≥2 | 5 (3) | 1 (4) | 4 (3) | 0 | 0 | 0 | |
| Number of post-op surgeries (minus chest closure), n (%) | 0.87 | ||||||
| 0 | 173 (90) | 24 (86) | 120 (90) | 21 (91) | 3 (100) | 5 (100) | |
| 1 | 16 (8) | 4 (14) | 10 (7) | 2 (9) | 0 | 0 | |
| ≥2 | 4 (2) | 0 | 4 (3) | 0 | 0 | 0 | |
| Subjects Discharged on Supplemental O2, n (%) | 30 (16) | 2 (7) | 26 (20) | 2 (9) | 0 | 0 | 0.37 |
| Oxygen Saturation at Discharge5 | |||||||
| Mean ± SD | 82.1 ± 5.0 | 81.7 ± 4.7 | 82.0 ± 5.2 | 82.5 ± 4.5 | 78.7 ± 6.1 | 85.8 ± 3.7 | 0.34 |
| Number of Discharge Medications | |||||||
| Median (IQR) | 4.0 (3, 5) | 3.5 (3, 5) | 4.0 (3, 5) | 4.0 (2, 5) | 2 (2, 2) | 4.0 (3, 4) | 0.29 |
| Any Adverse Events, n (%) | 52 (27) | 9 (32) | 35 (26) | 6 (26) | 1 (33) | 1 (20) | 0.94 |
| Types of Adverse Events3, n (%) | |||||||
| Cardiac General/Arrhythmia | 17 (9) | 2 (7) | 12 (9) | 1 (4) | 1 (33) | 1 (20) | 0.32 |
| Cardiac General | 9 (5) | 1 (4) | 6 (4) | 0 | 1 (33) | 1 (20) | 0.08 |
| Cardiac Arrhythmia | 8 (4) | 1 (4) | 6 (4) | 1 (4) | 0 | 0 | 1.00 |
| Infection | 17 (9) | 6 (21) | 9 (7) | 2 (9) | 0 | 0 | 0.17 |
| Pulmonary/Upper | |||||||
| Respiratory | 19 (10) | 2 (7) | 16 (12) | 1 (4) | 0 | 0 | 0.84 |
Note: Subjects who died or had heart transplants prior to stage 2 hospital discharge are excluded from the summary of hospital length of stay, total days in ICU, total days on ventilator, and days until last chest tube was removed.
BBDCPA=Bilateral bidirectional cavopulmonary anastomoses; BDCPA=Bidirectional cavopulmonary anastomosis; BCA=Bilateral cavopulmonary anastomoses
P-values assess the significance of the difference between types of SCPA and are based on the Kruskal-Wallace test for continuous data that are skewed, analysis of variance for other continuous data, and Fisher exact test for percentages.
Categories associated with at least 10 percent of the subjects are summarized.
Percentage is based on subjects with arrhythmia.
Includes only subjects who were not discharged on supplemental oxygen
Table 2.
Characteristics of Subjects Who Died and Subjects Who Underwent Heart Transplantation after SCPA
| Time to death/transplant (days) | On ACE inhibitor at pre-SCPA visit | Discharged after SCPA | Subject gender | Age at SCPA (months) | Weight-for-age z-score | Ventricular type | Oxygen saturation by pulse oximeter (%) | Pre-SCPA BNP (pg/ml) |
|---|---|---|---|---|---|---|---|---|
| Deaths | ||||||||
| 1 | No | No | Male | 5.0 | −4.12 | Right | 70 | 801.0 |
| 10 | Yes | No | Female | 4.9 | −1.37 | Right | 80 | 171.0 |
| 66 | No | Yes | Female | 5.7 | −2.96 | Left | 67 | 1190.0 |
| 95 | No | Yes | Male | 5.1 | −1.03 | Right | 72 | 141.0 |
| 144 | No | No | Male | 3.4 | −2.23 | Right | 80 | |
| 238 | No | Yes | Male | 8.2 | −2.69 | Right | 59 | |
| Heart Transplants | ||||||||
| 70 | Yes | No | Male | 3.0 | −1.73 | Right | 76 | 490.0 |
| 94 | Yes | Yes | Female | 5.9 | −2.37 | Right | 70 | 135.0 |
| 103 | No | Yes | Male | 3.3 | Right | 65 | ||
| 121 | No | Yes | Male | 2.4 | −2.83 | Right | 71 | 30.5 |
SCPA – superior cavopulmonary connection, BNP – brain natriuretic peptide, Cath=catheterization
Deaths and heart transplantations following SCPA
There were 3 deaths and 1 heart transplantation that occurred during the SCPA hospitalization. An additional 3 deaths and 3 heart transplantations occurred between SCPA hospital discharge and the final study visit at 14 months of age (Table 3). The incidence of death was 3% and heart transplantation was 2% by 14 months of age. Deaths occurred predominantly in subjects with morphologic right ventricles, though it should be noted that 72% of the cohort had a morphologic right ventricle. Subjects who died also generally had low weight-for-age z scores, with scores ranging from −4.2 to 1.0. Average weight-for-age z-score among subjects who survived SCPA was −1.61 (95% CI −1.77, −1.45).
Table 3.
Multivariate Models of Age at SCPA and SCPA Hospital Length of Stay
| Age at SCPA, months (R2=0.43): | Age at SCPA Median (IQR) | DF | Hazard Ratio (95% CI) | (N=169) P-value | |
|---|---|---|---|---|---|
| Center | -- | -- | 9 | -- | <0.001 |
| Race | White | 5.18 (4.2, 6.0) | 2 | 1.93 (1.2, 3.1) | 0.005 |
| Black | 6.03 (4.8, 7.4) | -- | |||
| Other | 4.52 (3.0, 5.7) | 3.03 (1.4, 6.4) | |||
| Age at Stage 1 (days), log transformation | 5.20 (4.2, 6.2) | 1 | 0.55 (0.41, 0.75) | <0.001 | |
| Norwood Procedure | Yes | 5.02 (4.1, 6.0) | 1 | 1.63 (1.1, 2.4) | 0.014 |
| No | 5.64 (5.0, 6.5) | -- | |||
| Inter-stage Pulmonary Adverse Event | Yes | 4.89 (3.9, 6.0) | 1 | 1.52 (1.1, 2.2) | 0.023 |
| No | 5.43 (4.6, 6.3) | -- | |||
| Inter-stage Cardiac Adverse Event | Yes | 4.74 (3.8, 5.7) | 1 | 2.31 (1.5, 3.5) | <0.001 |
| No | 5.31 (4.4, 6.3) | -- | |||
| Accessory Sources of Pulmonary Blood Flow pre-SCPA | Yes | 4.46 (4.0, 5.1) | 1 | 2.18 (1.2, 3.8) | 0.006 |
| No | 5.28 (4.4, 6.3) | -- |
| SCPA Hospital Length of Stay days natural log transformations (R2=0.33) | Predictor Mean± SD | DF | Slope (95% CI) | (N=170) P-value |
|---|---|---|---|---|
| Site | -- | 9 | -- | 0.10 |
| Number of Days on Ventilator during Palliative Surgery (square root) | 2.60 ± 1.05 | 1 | 0.13 (0.05, 0.21) | 0.003 |
| Weight-for-age z-score at Pre-SCPA | −1.61 ± 1.15 | 1 | −0.11 (−0.19, −0.04) | 0.003 |
| Pre-SCPA end systolic volume BSA-adjusted z-score | 3.37 ± 3.79 | 1 | 0.039 (0.017, 0.060) | <0.001 |
| Age at SCPA, log transformation, days | 5.07 ± 0.30 | 1 | −0.49 (−0.79, −0.18) | 0.002 |
| Number of Cardiac Surgical Procedures Concurrent with SCPA | 1.47 ± 1.12 | 1 | 0.16 (0.07, 0.25) | <0.001 |
Note: Hazard ratios and slopes associated with continuous variables represent the effect of a 1-unit increase of the continuous variable on the outcome. Among categorical variables, the reference group for hazard ratios is shown as a dash.
Age at SCPA
Median age at SCPA was 5.2 months (IQR 4.2, 6.2). The Cox proportional hazards regression model (N = 169, R2 = 0.43) identified independent associations of younger age at SCPA with the following factors: center, non-black race, younger age at neonatal palliative surgery, the performance of the Norwood procedure, time-related occurrence of interstage pulmonary or cardiac adverse events, and the presence of accessory pulmonary blood flow at pre-SCPA cardiac catheterization (Table 4).
Table 4.
Multivariate Models of Post-SCPA Unintended Intervention, Days on Ventilator, and Days on Chest Tube
| Post-SCPA Unintended Intervention (c-statistic=0.86) | No Intervention (N=151) | Intervention (N=21) | DF | Odds Ratio (95%CI) | P-value |
|---|---|---|---|---|---|
| Site | -- | -- | 9 | -- | 0.37 |
| Weight-for-age z-score | −1.52 ± 1.13 | −2.38 ± 1.13 | 1 | 0.50 (0.30, 0.83) | 0.007 |
| BSA-adjusted mass z-score, median cut-off | 1 | 0.013 | |||
| <3.94 | 83 (95.4) | 4 (4.6) | -- | ||
| ≥3.94 | 68 (80.0) | 17 (20.0) | 3.92 (1.34, 11.51) |
| Total Days on Ventilator (c-statistic=0.81) | ≤2 days (N=126) | >2 days (N=38) | DF | Odds Ratio (95%CI) | P-value |
|---|---|---|---|---|---|
| Site | -- | -- | 9 | -- | 0.48 |
| BNP (pg/mL), upper tertile | 1 | 0.005 | |||
| ≤135 | 91 (82.0) | 20 (18.0) | -- | ||
| >135 | 35 (66.0) | 18 (34.0) | 3.46 (1.45, 8.28) | ||
| Age at SCPA, lower tertile, days | 1 | 0.005 | |||
| >140 | 93 (81.6) | 21 (18.4) | -- | ||
| ≤140 | 33 (66.0) | 17 (34.0) | 3.59 (1.46, 8.85) | ||
| Number of Concurrent Cardiac Surgical Procedures | 1.43 ± 1.22 | 1.77 ± 0.96 | 1 | 1.92 (1.23, 3.02) | 0.005 |
| Days with Chest Tube (c-statistic=0.84) | ≤5 days (N=148) | >5 days (N=26) | DF | Odds Ratio (95%CI) | P-value |
|---|---|---|---|---|---|
| Site | -- | -- | 9 | -- | 0.044 |
| Pre-SCPA Ejection Fraction (%) | 58.29 ± 9.44 | 52.53 ± 9.57 | 1 | 0.93 (0.89, 0.98) | 0.003 |
Note: Odds ratios and slopes associated with continuous variables represent the effect of a 1-unit increase of the continuous variable on the outcome. Among categorical variables, the reference group for odds ratios is shown as a dash.
Length of stay following SCPA
Median LOS was 7 days (IQR 6, 10). After adjusting for center differences, the multivariable linear regression model of log LOS (N = 170, R2 = 0.33) identified independent associations of longer LOS with the following factors: greater number of ventilator days during the neonatal palliative surgery, lower weight-for-age z-score at pre-SCPA clinic visit, higher end systolic volume z-score relative to body surface area (BSA) at pre-SCPA echocardiogram, younger age at SCPA (log), and greater number of concurrent cardiac procedures at time of SCPA (Table 4).
Occurrence of 1 or more unanticipated interventions during SCPA hospitalization
At least one unanticipated surgical or cardiac catheterization intervention occurred during SCPA hospitalization in 12% of subjects. The types and frequencies of interventions are listed in Table 6 (online). After adjusting for center, the multivariable logistic regression model (N=172, c-statistic = 0.86) identified that the occurrence of 1 or more unanticipated interventions was associated with lower weight-for-age z-score and higher ventricular mass z-score relative to BSA on the pre-SCPA echocardiogram (Table 5).
Table 5.
Types of unanticipated postoperative procedures
| Category | n |
|---|---|
| Cardiothoracic surgery | |
| Thoracostomy tube | 11 |
| Chest exploration | 4 |
| Central or Blalock-Taussig shunt | 3 |
| Device implantation | 2 |
| Diaphragm plication | 2 |
| Pulmonary artery plasty | 1 |
| Shunt revision/thrombectomy | 1 |
| Extracorporeal membrane oxygenator support | 1 |
| Mediastinal debridement | 1 |
| Right pneumonectomy | 1 |
| Heart transplant | 1 |
| Ligation of left aortopulmonary collateral | 1 |
| Coarctation repair | 1 |
| Repair bleeding chest tube, evacuate hemothorax | 1 |
| Cardiac catheterization | |
| Angioplasty | |
| Pulmonary artery | 4 |
| Aorta | 2 |
| Superior vena cava | 2 |
| Stent | |
| Superior vena cava | 1 |
| Pulmonary artery | 1 |
| Coil | |
| Artery | 5 |
| Vein | 1 |
| Unspecified | 1 |
| Noncardiac | |
| Extremity amputation | 2 |
| Extremity debridement | 2 |
| Extremity skin graft | 1 |
| Gastrostomy-tube and Nissen fundoplication | 1 |
Mechanical ventilation for >2 days after SCPA
Mechanical ventilation was used for >2 days following SCPA in 23% of subjects. After adjusting for center, the multivariable logistic regression model (N=164, c-statistic 0.81) identified independent associations of mechanical ventilation for >2 days with the following factors: <140 days (lower tertile) of age at SCPA, BNP >135 pg/mL, and a greater number of concurrent cardiac surgical procedures at the time of SCPA (Table 5).
Final chest tube removed at >5 days after SCPA
The final chest tube was removed at >5 days after SCPA in 14% of subjects. After adjusting for center, the multivariable logistic regression model (N=174, c-statistic = 0.84) identified an independent association of final chest tube removal at >5 days with lower pre-SCPA ventricular ejection fraction (Table 5).
Discussion
By assessing factors associated with age at SCPA and measures of surgical outcome, the results of this analysis highlight the continued controversy about optimal timing of SCPA while offering insight on factors associated with surgical outcome.
Mortality and Transplantation
Similar to previous reports, deaths were relatively uncommon following SCPA.(5, 15) When compared to neonatal palliations, the SCPA may be less of a physiologic stress, is performed in some cases without cardiopulmonary bypass, and generally results in more favorable postoperative hemodynamics. It is also possible that the highest risk subjects die before SCPA.(11)
Subject and SCPA Hospitalization Characteristics
With the exception of the proportion of subjects having aortopulmonary shunt takedown, no differences in subject or SCPA hospitalization characteristics were noted across SCPA subtypes. While the results of subgroup comparisons should be interpreted cautiously given the small number of subjects for some groups, these results do not identify significant differences in subjects undergoing bilateral SCPA (16) or a Kawashima procedure.(17, 18)
Age at SCPA
This analysis identified subject characteristics or events that might result in younger age at SCPA. The time-related occurrences of interstage cardiac and pulmonary adverse events were associated with earlier SCPA. Consistent with previous reports, events such as residual arch obstruction(19) or narrowing of the aortopulmonary shunt(20) likely trigger earlier SCPA to address specific causes of the adverse event and reduce the negative impact on single ventricle physiology. In contrast, the results did not confirm previously reported significant associations between age at SCPA and several factors considered potentially clinically relevant such as type of SCPA, oxygen saturation, presence of valve regurgitation (15), decreased ventricular ejection fraction and increased ventricular end-diastolic pressure. Subjects who had a younger age at neonatal palliation or a Norwood operation may undergo SCPA sooner due to concerns of volume overload from accessory pulmonary blood flow or risk of interstage death. The independent association of race and timing of SCPA is difficult to interpret using variables derived solely from this data set. Largely a social construct, race is used in health research as a surrogate for many unmeasured environmental and biological differences between individuals. Details of these factors were not available. However, the observation is not entirely surprising as racial disparities have been described in almost every area of health care including congenital heart disease.(21)
The timing of SCPA is a decision that balances clinical driving forces (e.g. hypoxemia), prior operative results and anticipated benefits of altered physiology. However, timing may also be influenced by the logistical challenges of busy clinical programs such as bed space, operating room time, and physician availability. The interpretation of this balance is likely different among centers as evidenced by the observation that younger age was independently associated with center after adjusting for measured subject factors. Similar center differences in surgical timing have been reported in the single ventricle population.(22) These results highlight the lack of consensus regarding the optimal timing of the SCPA in infants with single ventricle.
Length of stay following SCPA
Several of the factors associated with LOS could be considered more generalized measures of subject well-being prior to SCPA. Particularly, weight-for-age z-score, which was chosen as the primary endpoint for the main trial, reflects broadly on overall health in infants. Growth variables have previously been associated with single ventricle outcomes.(1, 22, 23) Ventricular end-systolic volume was one of several echocardiographic measures included in the statistical analyses and may represent a surrogate for heart failure. The number of days on ventilator at the time of the first neonatal surgery could represent underlying subject characteristics that complicate cardiothoracic surgery or some degree of perioperative lung injury that subsequently complicates the SCPA. Finally, while the reasons that a subject undergoes an earlier SCPA are complex, younger age at SCPA was associated with longer hospital stay even when adjusting for these available factors. Given the ongoing debate about optimal timing of SCPA,(3, 6) this observation is notable. While the benefits of post-SCPA physiology may include favorable ventricular remodeling, less severe heart failure with improved growth, and avoidance of interstage death, it is possible that the potential benefits of elective SCPA at a younger age may be offset by potential complications related to a complex post-SCPA clinical course.
Other clinical outcomes after SCPA – unanticipated interventions, mechanical ventilation, chest tube duration
Analysis of unanticipated postoperative procedures provides further insight on the nature of the complications that occur in association with SCPA. The most common procedure in this study was a thoracostomy tube and the remaining procedures were quite varied. For this reason, the multivariable model should be interpreted conservatively and reflects the generalized risk of unanticipated procedures rather than risk for any specific event or procedure.
The binary outcomes for mechanical ventilator days (>2) and chest tube days (>5) were designed to identify predictors of outcomes considered suboptimal. The mean total duration for both variables was similar to prior reports.(1, 5) Younger age, along with elevated BNP and multiple concurrent operative procedures, was predictive of prolonged ventilator course. Worse ventricular function was predictive of chest tube drainage which may be related to elevation in pulmonary venous pressure resulting from high ventricular filling pressures. While direct catheter measurement did not support this hypothesis, the data from invasive measurements were frequently missing and power may be limited.
Study limitations
The limitations of this study include possible biases introduced by the original study cohort. Missing data, particularly hemodynamics from cardiac catheterization, limited the inclusion of some variables in multivariable modeling. Adverse events were only collected after the subject enrolled into the trial. For subjects who were enrolled into the trial after discharge from the neonatal hospitalization, events that may have occurred after discharge and prior to enrollment are not captured in these analyses. Socioeconomic data were insufficient to aid in interpretation of race findings in the age at SCPA model. Finally, while this is a relatively large cohort of single ventricle subjects, the low prevalence of suboptimal SCPA outcomes may result in limited power to detect associations in the multivariable model results in Table 4.
Conclusions
This large multicenter cohort of infants with single ventricle identified factors associated with timing of SCPA and surgical outcome. The incidence of death or heart transplantation was 5% by 14 months of age. Younger age at SCPA was associated with case complexity and interim morbidities. In addition, significant practice variation in the age at SCPA among centers, independent of subject factors, highlights the controversy regarding optimal timing of SCPA. Factors associated with LOS and other measures of operative morbidity may offer insights for improving pre-SCPA care and surgical outcome. The risks and benefits of elective SCPA at younger age require further study.
Acknowledgments
Supported by U01 grants from the National Heart, Lung, and Blood Institute (HL068269, HL068270, HL068279, HL068281, HL068285, HL068292, HL068290, HL068288, HL085057) and the FDA Office of Orphan Products Development. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of NHLBI or NIH
Abbreviations
- SCPA
superior cavopulmonary anastomosis
- ISV
Infant single ventricle
- LOS
length of stay
APPENDIX 1: List of Predictors Explored
The following list presents all the potential predictors that were explored. Tertiles and median cut-offs were included in logistic regression analyses to understand any possible non-linear association. In addition, when associations appeared non-linear according to the tertiles, the lowest tertile or highest tertile may have been analyzed as a dichotomous predictor. In the analysis of length of stay, the associations of predictors with the outcome were explored using generalized additive model (GAM) regressions and plots.
Treatment
NITT Enalapril (ACE inhibitor) at Pre-SCPA visit
Percent Time on Enalapril or ACE inhibitor
Demographics and baseline characteristics
Birth weight (kg) (observed, tertiles, log transformed, square root transformed, <2500 g)
Gestational age (observed, <37 weeks vs. ≥37 weeks)
Gender
Race
Parent’s Highest Level of Education (1 df and 5 df)
Annual household income (1 df and 4 df)
Ventricular morphology (left vs right and mixed)
Systemic ventricular dysfunction at baseline
Details of the neonatal palliative surgery
Age at
Circulatory arrest time (observed, tertiles, above/below median)
Length of time on ventilator (observed, tertiles, above/below median, square root transformed)
Number of Interventional catheterizations During neonatal hospitalization after palliative surgery
Number of Other Surgeries after Palliative Surgery (collapsed to any other surgeries)
Pre-SCPA measures
Pre-SCPA weight-for-age z-score (observed, <-2 vs ≥-2, <-1 vs ≥-1, tertiles)
Pre-SCPA O2 saturation (observed, tertiles, above/below median)
Pre-SCPA BNP (observed, tertiles, low BNP, log transformed)
Pre-SCPA Catheterization
Mean systemic venous atrial pressure (observed, tertiles, above/below median)
Pulmonary capillary wedge pressure (observed, tertiles, above/below median)
End diastolic pressure (observed, tertiles, above/below median)
Mean pulmonary artery pressure (observed, tertiles, above/below median)
Transpulmonary gradient (observed, tertiles, above/below median)
Number of interventional procedures during the catheterization
Pre-SCPA Echocardiogram
Overall AV valve regurgitation
Ejection fraction (observed, >55% vs≤55%, z-score)
Mass to volume ratio (observed, tertiles, above/below median)
Mass to volume ratio z-score (observed, tertiles, above/below median)
Mass (observed, tertiles, above/below median)
Mass z-score (observed, tertiles, above/below median)
End diastolic volume (observed, tertiles, above/below median)
End diastolic volume z-score (observed, tertiles, above/below median)
End systolic volume (observed, tertiles, above/below median)
End systolic volume z-score (observed, tertiles, above/below median)
Pre-SCPA Feeding
Whether the subject was on solid food pre-SCPA (yes/no)
Total daily calories consumed (observed, tertiles, above/below median, log transformed)
Total calories consumed per day per kg (observed, tertiles, above/below median, log transformed)
Pre-SCPA feeding type
Tube fed Pre-SCPA (yes/no)
Interstage Adverse Events
The first serious adverse event
The first cardiac general or cardiac arrhythmia adverse event
The first pulmonary adverse event
Details of the SCPA
Age at SCPA (observed, tertiles, <140 days vs ≥140 days), for SCPA outcomes
Type of SCPA (bidirectional Glenn, hemi-Fontan, bilateral bidirectional Glenn, Kawashima)
Number of Cardiac Procedures Concurrent with SCPA
APPENDIX 2
National Heart, Lung, and Blood Institute: Gail Pearson, Victoria Pemberton, Mario Stylianou, Marsha Mathis*
Network Chair: Lynn Mahony, University of Texas Southwestern Medical Center
Data Coordinating Center: New England Research Institutes, Lynn Sleeper (PI), Steven Colan, Lisa Virzi*, Lisa Wruck*, Victor Zak, David F. Teitel
Core Clinical Site Investigators: Children’s Hospital Boston, Jane W. Newburger (PI), Roger Breitbart, Jami Levine, Ellen McGrath, Carolyn Dunbar-Masterson; Children’s Hospital of New York, Daphne Hsu* (Study Chair), William Hellenbrand (PI), Ashwin Prakash*, Seema Mital*, Darlene Servedio*; Children’s Hospital of Philadelphia, Victoria L. Vetter (PI), Chitra Ravishankar, Sarah Tabbutt*, Meryl Cohen, Katherine Lee, Marisa Nolan, Stephanie Piacentino, Michelle Toms; Cincinnati Children’s Medical Center, D. Woodrow Benson (PI), Catherine Dent Krawczeski, Lois Bogenschutz, Teresa Barnard, Steven Schwartz*, David Nelson; North Carolina Consortium: Duke University, East Carolina University, Wake Forest University, Page A. W. Anderson (PI) – deceased, Jennifer Li (PI), Wesley Covitz, Kari Crawford*, Michael Hines, James Jaggers*, Theodore Koutlas, Charlie Sang, Jr, Lori Jo Sutton, Mingfen Xu; Medical University of South Carolina, J. Philip Saul (PI), Andrew Atz, Girish Shirali, Eric M. Graham, Teresa Atz; Primary Children’s Medical Center and the University of Utah, Salt Lake City, Utah, L. LuAnn Minich (PI), John A. Hawkins – deceased, Richard V. Williams, Linda M. Lambert, Marian E. Shearrow; Hospital for Sick Children, Toronto, Brian McCrindle (PI), Elizabeth Radojewski, Nancy Slater, Svetlana Khaikin, Susan McIntyre.
Auxiliary Sites: Children’s Hospital of Wisconsin, Nancy Ghanayem, Kathy Mussatto, Michele Frommelt, Lisa Young-Borkowski; University of Michigan, Albert Rocchini, Laurie Rodgers-Augustyniak
Echocardiography Core Laboratory: Children’s Hospital Boston: Steven Colan, Renee Margossian
Genetics Core Laboratory: Children’s Hospital of New York: Wendy Chung, Liyong Deng, Patricia Lanzano
Protocol Review Committee: Michael Artman, Chair; Judith Massicot-Fisher, Executive Secretary; Timothy Feltes, Julie Johnson, Thomas Klitzner, Jeffrey Krischer, G. Paul Matherne
Data and Safety Monitoring Board: John Kugler, Chair; Rae-Ellen Kavey, Executive Secretary; David J. Driscoll, Mark Galantowicz, Sally A. Hunsberger, Thomas J. Knight, Holly Taylor, Catherine L. Webb
Footnotes
no longer at the institution listed
ClinicalTrials.gov number: NCT00113087
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Anderson JB, Beekman RH, 3rd, Border WL, Kalkwarf HJ, Khoury PR, Uzark K, et al. Lower weight-forage z score adversely affects hospital length of stay after the bidirectional Glenn procedure in 100 infants with a single ventricle. J Thorac Cardiovasc Surg. 2009 Aug;138(2):397–404. e1. doi: 10.1016/j.jtcvs.2009.02.033. [DOI] [PubMed] [Google Scholar]
- 2.Ashburn DA, Blackstone EH, Wells WJ, Jonas RA, Pigula FA, Manning PB, et al. Determinants of mortality and type of repair in neonates with pulmonary atresia and intact ventricular septum. J Thorac Cardiovasc Surg. 2004 Apr;127(4):1000–7. doi: 10.1016/j.jtcvs.2003.11.057. discussion 7–8. [DOI] [PubMed] [Google Scholar]
- 3.Ghanayem NS, Tweddell JS, Hoffman GM, Mussatto K, Jaquiss RD. Optimal timing of the second stage of palliation for hypoplastic left heart syndrome facilitated through home monitoring, and the results of early cavopulmonary anastomosis. Cardiol Young. 2006 Feb;16(Suppl 1):61–6. doi: 10.1017/S1047951105002349. [DOI] [PubMed] [Google Scholar]
- 4.Jaquiss RD, Siehr SL, Ghanayem NS, Hoffman GM, Fedderly RT, Cava JR, et al. Early cavopulmonary anastomosis after Norwood procedure results in excellent Fontan outcome. Ann Thorac Surg. 2006 Oct;82(4):1260–5. doi: 10.1016/j.athoracsur.2006.04.095. discussion 5–6. [DOI] [PubMed] [Google Scholar]
- 5.Kogon BE, Plattner C, Leong T, Simsic J, Kirshbom PM, Kanter KR. The bidirectional Glenn operation: a risk factor analysis for morbidity and mortality. J Thorac Cardiovasc Surg. 2008 Nov;136(5):1237–42. doi: 10.1016/j.jtcvs.2008.05.017. [DOI] [PubMed] [Google Scholar]
- 6.Petrucci O, Khoury PR, Manning PB, Eghtesady P. Outcomes of the bidirectional Glenn procedure in patients less than 3 months of age. J Thorac Cardiovasc Surg. 2009 Nov 10; doi: 10.1016/j.jtcvs.2009.08.025. [DOI] [PubMed] [Google Scholar]
- 7.Slavik Z, Lamb RK, Webber SA, Devlin AM, Keeton BR, Monro JL, et al. Bidirectional superior cavopulmonary anastomosis: how young is too young? Heart. 1996 Jan;75(1):78–82. doi: 10.1136/hrt.75.1.78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kaltman JR, Schramm C, Pearson GD. The National Heart, Lung, and Blood Institute bench to bassinet Program: a new paradigm for translational research. J Am Coll Cardiol. 2010 Mar 23;55(12):1262–5. doi: 10.1016/j.jacc.2009.11.055. [DOI] [PubMed] [Google Scholar]
- 9.Wernovsky G, Ghanayem N, Ohye RG, Bacha EA, Jacobs JP, Gaynor JW, et al. Hypoplastic left heart syndrome: consensus and controversies in 2007. Cardiol Young. 2007 Sep;17(Suppl 2):75–86. doi: 10.1017/S1047951107001187. [DOI] [PubMed] [Google Scholar]
- 10.Ohye RG, Goldberg CS, Donohue J, Hirsch JC, Gaies M, Jacobs ML, et al. The quest to optimize neurodevelopmental outcomes in neonatal arch reconstruction: the perfusion techniques we use and why we believe in them. J Thorac Cardiovasc Surg. 2009 Apr;137(4):803–6. doi: 10.1016/j.jtcvs.2008.10.046. [DOI] [PubMed] [Google Scholar]
- 11.Hsu DT, Zak V, Mahony L, Sleeper LA, Atz AM, Levine JC, et al. Enalapril in infants with single ventricle: results of a multicenter randomized trial. Circulation. 2010 Jul 27;122(4):333–40. doi: 10.1161/CIRCULATIONAHA.109.927988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hsu DT, Mital S, Ravishankar C, Margossian R, Li JS, Sleeper LA, et al. Rationale and design of a trial of angiotensin-converting enzyme inhibition in infants with single ventricle. Am Heart J. 2009 Jan;157(1):37–45. doi: 10.1016/j.ahj.2008.08.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.National Cancer Institute (U.S.) Common terminology criteria for adverse events (CTCAE) Bethesda, Md: U.S. Dept. of Health and Human Services, National Institutes of Health, National Cancer Institute; 2009. Rev. ed. [Google Scholar]
- 14.Sluysmans T, Colan SD. Theoretical and empirical derivation of cardiovascular allometric relationships in children. J Appl Physiol. 2005 Aug;99(2):445–57. doi: 10.1152/japplphysiol.01144.2004. [DOI] [PubMed] [Google Scholar]
- 15.Scheurer MA, Hill EG, Vasuki N, Maurer S, Graham EM, Bandisode V, et al. Survival after bidirectional cavopulmonary anastomosis: analysis of preoperative risk factors. J Thorac Cardiovasc Surg. 2007 Jul;134(1):82–9. 9 e1–2. doi: 10.1016/j.jtcvs.2007.02.017. [DOI] [PubMed] [Google Scholar]
- 16.Iyer GK, Van Arsdell GS, Dicke FP, McCrindle BW, Coles JG, Williams WG. Are bilateral superior vena cavae a risk factor for single ventricle palliation? Ann Thorac Surg. 2000 Sep;70(3):711–6. doi: 10.1016/s0003-4975(00)01627-1. [DOI] [PubMed] [Google Scholar]
- 17.Brown JW, Ruzmetov M, Vijay P, Rodefeld MD, Turrentine MW. Pulmonary arteriovenous malformations in children after the Kawashima operation. Ann Thorac Surg. 2005 Nov;80(5):1592–6. doi: 10.1016/j.athoracsur.2005.04.043. [DOI] [PubMed] [Google Scholar]
- 18.Nath DS, Carden AJ, Nussbaum DP, Shin AJ, Khemani RG, Starnes VA, et al. Can the Kawashima procedure be performed in younger patients? Ann Thorac Surg. 2009 Aug;88(2):581–6. doi: 10.1016/j.athoracsur.2009.04.045. discussion 6–7. [DOI] [PubMed] [Google Scholar]
- 19.Chessa M, Dindar A, Vettukattil JJ, Stumper O, Wright JG, Silove ED, et al. Balloon angioplasty in infants with aortic obstruction after the modified stage I Norwood procedure. Am Heart J. 2000 Aug;140(2):227–31. doi: 10.1067/mhj.2000.108238. [DOI] [PubMed] [Google Scholar]
- 20.Wang JK, Wu MH, Chang CI, Chiu IS, Lue HC. Balloon angioplasty for obstructed modified systemic-pulmonary artery shunts and pulmonary artery stenoses. J Am Coll Cardiol. 2001 Mar 1;37(3):940–7. doi: 10.1016/s0735-1097(00)01194-3. [DOI] [PubMed] [Google Scholar]
- 21.Nembhard WN, Salemi JL, Ethen MK, Fixler DE, Dimaggio A, Canfield MA. Racial/Ethnic disparities in risk of early childhood mortality among children with congenital heart defects. Pediatrics. 2011 May;127(5):e1128–38. doi: 10.1542/peds.2010-2702. [DOI] [PubMed] [Google Scholar]
- 22.Wallace MC, Jaggers J, Li JS, Jacobs ML, Jacobs JP, Benjamin DK, et al. Center variation in patient age and weight at Fontan operation and impact on postoperative outcomes. Ann Thorac Surg. 2011 May;91(5):1445–52. doi: 10.1016/j.athoracsur.2010.11.064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Anderson JB, Kalkwarf HJ, Kehl JE, Eghtesady P, Marino BS. Low weight-for-age z-score and infection risk after the Fontan procedure. Ann Thorac Surg. 2011 May;91(5):1460–6. doi: 10.1016/j.athoracsur.2011.01.020. [DOI] [PubMed] [Google Scholar]