- Email: [email protected]
Survival in Children With Down Syndrome Undergoing Single-Ventricle Palliation
Survival in Children With Down Syndrome Undergoing Single-Ventricle Palliation John L. Colquitt, MD, Shaine A. Morris, MD, MPH, Susan W. Denfield, MD, Charles D. Fraser, MD, Yunfei Wang, PhD, and W. Buck Kyle, MD Division of Cardiology, Department of Pediatrics, and Division of Congenital Heart Surgery, Michael E. DeBakey Department of Surgery, Baylor College of Medicine and Texas Children’s Hospital, Houston, Texas
Background. We describe survival in patients with Down syndrome (DS) with single-ventricle anatomy and palliation and characterize risk factors associated with mortality. Methods. All patients with DS and single-ventricle anatomy documented in the electronic medical record at our institution from January 1, 1992, to May 1, 2014, were compared with patients with unbalanced atrioventricular septal defects and single-ventricle anatomy, without DS or heterotaxy, during the same period. Survival analysis was performed to evaluate factors associated with mortality, including the presence of DS. Results. There were 28 patients with DS and 30 patients without DS. One-year survival with DS was 75% (95% CI: 55% to 87%); 5-year survival was 61% (95% CI: 40% to 76%). All DS deaths except one occurred before 2 years of age. One-year non-DS survival was 93% (95% CI: 76% to 98%); 5-year survival was 85% (95% CI: 64% to 94%). Factors associated with
death by univariable analysis included DS (p [ 0.04), pulmonary vascular resistance (PVR) of at least 3 Wood units 3 meter2 (WUm2) in the first year of life (p [ 0.03), and moderate-to-severe atrioventricular valve regurgitation (p [ 0.1). In combined analysis, when accounting for PVR of at least 3 WUm2 (hazard ratio [HR] 9.8, 95% CI: 1.1 to 83.5, p [ 0.04), DS was not associated with increased mortality (HR 1.5, 95% CI: 0.3 to 7.8, p [ 0.66). No patient with DS with PVR less than 3 WUm2 died. Conclusions. Children with DS and single-ventricle anatomy have excellent survival when PVR is less than 3 WUm2 in the first year of life, with minimal mortality beyond 2 years of age. When accounting for PVR, DS alone is not associated with increased mortality in patients with single-ventricle anatomy.
P
patients with and without DS who undergo SVP, to characterize risk factors associated with mortality in DS and SVP, and to investigate whether DS alone is a risk factor for death.
atients with Down syndrome (DS) and congenital heart disease who undergo single-ventricle palliation (SVP) represent a population considered to be at high risk for morbidity and mortality compared with their euploidic peers. Low transpulmonary pressures are necessary for successful palliation, but children with DS are inherently predisposed to higher pulmonary vascular resistance (PVR) [1–3]. Recent large administrative database studies have shown increased hospital deaths after staged palliation procedures in patients with DS compared with patients without DS [4, 5]. Other studies have reported on mortality in DS after total cavopulmonary connection (TCPC) [2, 6–9], with inconsistent findings about mortality risk compared with patients without DS. Overall, little has been published on the clinical and surgical outcomes of patients with DS, particularly before TCPC. The purpose of this study is to describe survival in patients with DS and single-ventricle anatomy, to compare survival in
Accepted for publication Nov 17, 2015. Address correspondence to Dr Kyle, Texas Children’s Hospital, Baylor College of Medicine, Department of Pediatrics, Division of Pediatric Cardiology, 6621 Fannin, MC# 19345-C, Houston, TX 77030; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Patients and Methods This study was approved by the Institutional Review Board at Baylor College of Medicine. Patients cared for between January 1, 1992, and May 1, 2014, were identified with institutional electronic databases. Search criteria included diagnosis of DS and single-ventricle anatomy. Single-ventricle anatomy was defined as the inability to surgically separate pulmonary and systemic cardiac output. For patients who underwent a surgical procedure, this included pulmonary artery banding, pulmonary artery banding with superior cavopulmonary connection (SCPC; also known as pulsatile Glenn), and intact pulmonary artery with SCPC (also known as one-and-a-half repair). In most patients, suitability for septation was determined by echocardiogram; however, in some, it was determined on intracardiac inspection during the operation. Patients without DS who had an unbalanced atrioventricular septal defect (AVSD) considered unsuitable for septation were selected for comparison with the 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.11.047
2
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
use of the same search criteria. Patients with unbalanced AVSD were chosen for comparison, given that this lesion was present in most of the DS cohort. Patients with heterotaxy complex (isomerism) were excluded because of a known increased risk of death [10, 11]. Clinical diagnosis, demographic characteristics, gestational age, dominant ventricle (morphologic left versus right), age at surgical procedure, and type of pre-SCPC surgical procedure (pulmonary artery band, systemic-to-pulmonary artery shunt, or other) were recorded. Preterm was defined as birth less than 37 weeks’ gestation. Preoperative data collected included degree of atrioventricular valve regurgitation, mean pulmonary artery pressure, ventricular end-diastolic pressure, and calculated PVR as obtained by echocardiography and cardiac catheterization. The degree of atrioventricular valve regurgitation was estimated qualitatively and was divided into 1, none to mild, and 2, moderate to severe. Highest PVR in the first year of life was collected when available. PVR is reported in Wood unit meter2 (WUm2). Pre-SCPC PVR was defined as the PVR measured on the catheterization most closely preceding the SCPC surgical procedure.
Statistical Methods Wilcoxon rank sum and Pearson c2 tests were used to compare continuous and categorical variables, respectively, and Cochran-Armitage trend test was applied for ordinal variables. Cardiac survival (from birth to death or heart transplantation) was estimated with the KaplanMeier method. Birth was time zero, and all living patients without transplantation were censored at the date of last contact. Kaplan-Meier analysis was first used, including patients with and without DS, to perform univariable analysis to determine which factors were associated with mortality. A receiver operating characteristic curve analysis was created to identify a threshold value of PVR that maximized sensitivity and specificity for mortality to delineate a clinically useful cutoff. An exploratory multivariable Cox regression model was then created, initially including all covariates associated with mortality with p less than 0.2 by univariable analysis. Given the small dataset, backward elimination was used to discard covariates with p values greater than 0.05, retaining DS in the model to assess its effect estimate. Given high collinearity between PVR and mean pulmonary arterial pressure, only PVR was used in the model. A p value less than 0.05 was considered statistically significant. Analysis was performed with SAS version 9.4 (SAS Institute, Cary, NC).
Results Twenty-eight patients with DS and single-ventricle anatomy and 30 patients without DS with unbalanced AVSD were included. Patient characteristics are listed in Table 1. Median age at last follow-up was approximately 5 years in both groups (range, 2 months to 33 years in both). Statistically significant differences between the groups included older median age at SCPC (p ¼ 0.03), higher preSCPC PVR (p ¼ 0.02), and higher pre-SCPC mean
Ann Thorac Surg 2016;-:-–-
pulmonary artery pressure (p ¼ 0.01) in the DS group. The DS group was evenly divided between morphologic left (50%) and right (50%) ventricles, whereas 70% of the non-DS group had a morphologic right ventricle (p < 0.01). The majority of patients with DS had an unbalanced AVSD (26 of 28, 93%); one patient had hypoplastic left heart syndrome and another had tricuspid atresia. The clinical course of patients with DS is shown in Figure 1. The majority of patients were initially palliated with a pulmonary artery band (22 of 28, 79%), including 5 patients with a concomitant arch repair. Three had a Blalock-Taussig shunt (3 of 28, 11%), 1 of whom underwent a Norwood procedure. Of the remaining 3, 1 died before any operation and the other 2 underwent SCPC as their first palliation. SCPC was performed in 16 patients with DS (16 of 28, 57%) at a median age of 1.1 years (range, 3 months to 13.4 years). In contrast, median age at SCPC in patients without DS was 6 months (range, 3 months to 6.6 years; p ¼ 0.03). The median SCPC date for patients with DS and patients without DS was September 12, 2006 (range, June 23, 1997 to March 8, 2011) and March 18, 2008 (range, February 1, 1992 to April 14, 2014), respectively. Of the DS group, 8 remain with only SCPC, 3 have undergone TCPC, 1 has undergone a one-and-a-half ventricle repair, 2 died before another operation, and 2 underwent SCPC takedown for failing physiology. PVR before takedown in those 2 patients was 5.4 and 3.4. Data, including current management plan for patients with DS with SCPC, are found in Table 2. Presently, 3 patients who underwent SCPC are considered candidates for TCPC and are awaiting an operation. Two patients with DS underwent single-stage TCPC after pulmonary artery banding (ages at TCPC were 1.9 and 3.5 years). Thus, a total of 5 patients with DS achieved TCPC (18%; 5 of 28), at a median age of 3.5 years (range, 1.9 to 4.8 years). The median TCPC date for patients with DS and patients without DS was July 9, 2007 (range, June 23, 1997, to April 28, 2009) and September 19, 2009 (range, January 18, 1995, to June 17, 2013), respectively. The patients with DS with hypoplastic left heart syndrome and tricuspid atresia both successfully achieved TCPC. Four of the 5 are alive (age range at follow up, 5 to 11 years). TCPC was achieved in 16 patients without DS (53%). Sixteen patients with DS (57%) and 15 patients without DS (50%) underwent cardiac catheterization in the first year of life. Hemodynamic data are listed in Table 1. In patients with DS, 1-year Kaplan-Meier survival was 75% (95% CI: 55% to 87%); 5-year survival was 61% (95% CI: 40% to 76%). All deaths occurred before 2 years of age, except for 1 patient, which occurred at age 18 years. For patients without DS, 1-year Kaplan-Meier survival was 93% (95% CI: 76% to 98%); 5-year survival was 85% (95% CI: 64% to 94%; univariable comparison of overall Kaplan-Meier survival in DS and non-DS p ¼ 0.03). Median age at death was 11 months (range, 2 months to 18 years) for patients with DS and 2.3 years (range, 2 months to 5.7 years) for patients without DS. No patients with DS received a transplant. One patient without DS received a transplant; another is currently on the waiting list. Data
Ann Thorac Surg 2016;-:-–-
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
3
Table 1. Patient Characteristics Variable Male sex, n (%) Race/ethnicity, n (%) White Hispanic Black Other Lesion, n (%) Unbalanced AVSD HLHS Tricuspid atresia Systemic ventricle, n (%) Morphologic left Morphologic right Indeterminate Gestational age, n (%) Preterm Term No data Age at last follow-up, years, median (range) PVR 3 WUm2 in first year of life, n (%) Yes No No data Highest PVR in first year, WUm2, median (range) [n] First surgical procedure, n (%) PAB BT shunt SCPC Norwood palliation Other No data None Pre-SCPC AV valve regurgitation, n (%) None-mild Moderate-severe No data Age at pre-SCPC catheterization, years, median (range) [n] PVR pre-SCPC, WUm2, median (range) [n] Mean PAP pre-SCPC, mm Hg, median (range) [n] Age at SCPC, years, median (range) [n] LOS at SCPC, days, median (range) [n] Pre-TCPC AV valve regurgitation, n (%) None-mild Moderate-severe No TCPC PVR pre-TCPC, WUm2, median (range) [n] Mean PAP pre-TCPC, mm Hg, median (range) [n] Age at TCPC, years, median (range) [n] LOS at TCPC in days, median (range) [n] Age at death or heart transplantation, years, median (range) [n]
Non-Down Syndrome (n ¼ 30) 13 (43) 16 6 6 2
Down Syndrome (n ¼ 28) 7 (25)
(53) (20) (20) (7)
14 6 6 2
p Value 0.14
(50) (21) (21) (7)
1.00
30 (100) . .
26 (93) 1 (3) 1 (3)
0.33
2 (7) 21 (70) 7 (23)
13 (46) 13 (46) 2 (7)
<0.01
6 16 8 5.2
(20) (73) (27) (0.2–33.6)
4 16 8 5.1
(14) (57) (29) (0.2–33.1)
0.58
3 11 16 1.9
(10) (37) (53) (0.8–8.4) [14]
9 7 12 3.1
(32) (25) (43) (0.4–12.8) [16]
0.05
12 7 1 7 1 2
(40) (23) (3) (23) (3) (7)
22 2 2 1
(79) (7) (7) (3) . . 1 (3)
. 11 17 2 0.5 1.7 13 0.5 10
(37) (57) (7) (0.2–1.3) [15] (0.6–5.8) [15] (6–23.5) [15] (0.3–6.6) [26] (5–95) [22]
16 10 2 0.6 2.7 18 1.1 10
(57) (36) (7) (0.2–18.1) [16] (0.4–12.8) [15] (9–55) [15] (0.3–13.4) [19] (5–51) [16]
6 13 11 1.5 12 4.2 14 2.3
(20) (43) (37) (0.4–5.7) [18] (7–16) [18] (1.4–23.3) [16] (6–24) [16] (0.2–5.7) [5]
11 2 15 2.3 12.3 3.6 11 0.9
(39) (7) (54) (0.7–4.7) [12] (7–16) [12] (1.9–4.8) [5] (7–57) [5] (0.2–18) [11]
AV ¼ atrioventricular; AVSD ¼ atrioventricular septal defect; BT ¼ Blalock-Taussig; length of stay; PAB ¼ pulmonary artery band; PAP ¼ pulmonary artery pressure; superior cavopulmonary connection; TCPC ¼ total cavopulmonary connection.
HLHS ¼ hypoplastic left heart syndrome; PVR ¼ pulmonary vascular resistance;
0.3
0.08 0.02
0.10
0.55 0.02 0.01 0.03 0.63 <0.01
0.25 0.48 0.21 0.5 0.46 LOS ¼ SCPC ¼
4
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
Ann Thorac Surg 2016;-:-–-
Fig 1. Management and outcomes in children with Down syndrome and FSV. 1Included 1 Norwood arch reconstruction. 2One patient still has PAB; SCPC delayed because of PVR ¼ 10; group also includes five arch repairs. 3Attempted biventricular repair was unsuccessful and was taken down to PAB with subsequent demise. 4Takedown performed for failing SCPC physiology; 1 of 2 patients died. 5Developed progressive cyanosis/ exercise intolerance many years after PAB; could not septate, instead underwent palliative shunting. *Three patients who underwent SCPC are considered candidates for and await TCPC; if successful this percentage would increase to 32%. (BiV ¼ biventricular; BT ¼ Blalock-Taussig; FSV ¼ functional single ventricle; PAB ¼ pulmonary artery band; SCPC ¼ superior cavopulmonary connection; TCPC ¼ total cavopulmonary connection.)
related to deaths in the patients with DS can be found in Table 2. A receiver operating characteristic curve was created for PVR with the outcome of death or transplantation in all patients to determine a clinically useful cutoff for PVR, with the goal of maximizing sensitivity and specificity. A value of 2.8 WUm2 or more for PVR in the first year of life resulted, with a sensitivity of 86% and specificity of 74% (Fig 2). For clinical purposes, a value of 3 WUm2 was used (same sensitivity and specificity), and subsequent analyses that include PVR used values less than 3 WUm2 and 3 WUm2 or more. On univariable analysis, factors associated with death or transplantation included the presence of DS, PVR of at least 3 WUm2 in the first year of life, higher mean pulmonary arterial pressure before SCPC, worse atrioventricular valve regurgitation before SCPC, and higher ventricular end-diastolic pressure before SCPC (Table 3). No patients with DS and a PVR less than 3 WUm2 died (Fig 3). By univariable analysis, the hazard ratio (HR) for mortality for PVR of at least 3 WUm2 in the first year of life was 10.6 (95% CI: 1.3 to 87.9). By multivariable analysis (Table 3), the only variable that remained significantly associated with mortality was
PVR of at least 3 WUm2 in the first year of life (HR 9.75, 95% CI: 1.14 to 83.45). DS itself was not associated with increased mortality (HR 1.5, 95% CI: 0.3 to 7.8, p ¼0.66).
Comment To our knowledge, this is the largest single-center report of outcomes in patients with DS and single-ventricle anatomy. Our study confirms that SVP in patients with DS is associated with appreciable mortality; however, this appears to be accounted for by elevated PVR, because patients with DS and PVR less than 3 WUm2 had 100% survival in our study. PVR may be useful in risk stratifying this cohort to identify those at highest risk. Elevated PVR is a well-documented risk factor for poor outcomes with SVP [12, 13]. In a 2012 study of 227 patients with variable single-ventricle disorders, Alsoufi and colleagues [12] showed that risk factors for death before TCPC included a PVR more than 3 WUm2. Our data support this association in the DS population, with a HR of 9.8 when PVR was at least 3 WUm2 in the first year of life. As discussed in the Results section, elevated PVR (3 WUm2) was noted in patients who did not progress in
Ann Thorac Surg 2016;-:-–-
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
5
Table 2. Patients With Down Syndrome Age Current patients with SCPC 7y 7y 9y 5y 8y 5y 17 y 18 y
Most Recent PVR, WUm2
Comments
<1 2.6 <1 2.4 4.7 2.3
Considered a TCPC candidate Considered a TCPC candidate Considered a TCPC candidate Complicated post-SCPC course; TCPC candidacy in question Complicated post-SCPC course; TCPC candidacy in question Concerning pulmonary-to-systemic venous collateral; TCPC candidacy in question Parental choice to defer catheterization and operation Severe developmental delay, not ambulatory; not a TCPC candidate
No data 2.3 Most Recent PVR, WUm2
Comments
Current patients with definitive palliation 8y 11 y 9y
1.3 2.5 1.9
5y
2.3
5y
1.6
HLHS; nonfenestrated extracardiac conduit TCPC at 2.5 years Tricuspid atresia; fenestrated extracardiac TCPC at 4.3 years RV dominant AVSD; nonfenestrated extracardiac TCPC at 4.8 years LV dominant AVSD; fenestrated lateral tunnel TCPC at 3.5 years LV dominant AVSD; one-and-a-half ventricle repair at 4 years
Age
Age at Death Deaths 2 mo 2 mo 4 mo 5 mo 8 mo 10 mo 14 mo 19 mo 20 mo 24 mo 18 y Childhood
PVR, WUm2
Anatomy and Surgical History
Cause of Death
7.3 No data No data 12.8 No data 3.9 No data 3.5 3 3.1 7.2 No data
RV dominant AVSD, coarctation s/p PAB and arch repair RV dominant AVSD, type A interrupted arch s/p PAB, arch repair RV dominant AVSD, s/p failed attempt at 2V repair, PAB, ECMO RV dominant AVSD, none LV dominant AVSD, pulmonary atresia s/p BT shunt RV dominant AVSD, arch hypoplasia s/p PAB, arch repair; SCPC LV dominant AVSD s/p PAB; SCPC; ECMO LV dominant AVSD s/p PAB; SCPC; SCPC takedown to PAB RV dominant AVSD s/p PAB; SCPC; BT shunt LV dominant AVSD s/p PAB; single-stage TCPC LV dominant AVSD s/p PAB RV dominant AVSD, arch hypoplasia s/p PAB, arch repair
VT, cardiac arrest Heart failure Heart failure Heart failure Cardiac arrest Cardiac arrest Heart failure Heart failure Heart failure Heart failure No data No data
AVSD ¼ atrioventricular septal defect; BT ¼ Blalock-Taussig; ECMO ¼ extracorporeal membrane oxygenation; HLHS ¼ hypoplastic left heart syndrome; LV ¼ left ventricle; PAB ¼ pulmonary artery band; PVR ¼ pulmonary vascular resistance; RV ¼ right ventricle; SCPC ¼ superior cavopulmonary connection; s/p ¼ status post; TCPC ¼ total cavopulmonary connection; VT ¼ ventricular tachycardia; 2V ¼ biventricular.
their palliation and in patients who failed SCPC. Even in patients with DS and PVR 3 WUm2 or more, mortality beyond 2 years of age was minimal in our study. Previous literature has focused on TCPC outcomes in DS. In a 2010 study, Gupta-Malhotra and colleagues [7] reported an early mortality after TCPC of 35% and found DS to be an independent variable associated with a significantly higher risk of mortality in the early postoperative period. In contrast, in 2013, Furukawa and colleagues [6] concluded that the overall mortality rate of patients with DS undergoing TCPC was lower (12.5%), with no significant difference from that of children without DS.
Evident in this and other studies is that TCPC is infrequently attempted in patients with DS [4, 5, 7]. In a 2010 study, there were only 17 TCPC operations among 4,350 patients with DS in The Society of Thoracic Surgeons Congenital Heart Surgery Database [5]. In the referenced single-institution studies [2, 6, 8, 9], the reported TCPC numbers were 4, 8, 3, and 2, respectively. Including the one patient who underwent a one-and-ahalf ventricle repair, the number achieving what would be considered definitive palliation in our cohort was 6 (6 of 28, 21%). The group of incompletely palliated patients with elevated PVR presents a clinical dilemma (see Table 2 for comments about patients who underwent
6
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
Fig 2. Receiver operating characteristic curve for PVR in the first year of life and death/cardiac transplantation. A value of 3 WUm2 for PVR had a sensitivity of 86% and specificity of 74%. (AUC ¼ area under the curve; PVR ¼ pulmonary vascular resistance; WUm2 ¼ Wood unit m2.)
SCPC not offered TCPC). The addition of pulmonary vasodilators may be reasonable and was used in some of our patients. Surgical relief of upper airway obstruction might be entertained [14].
Ann Thorac Surg 2016;-:-–-
Given the poor outcomes reported in DS and SVP, there has been some recent consideration of aggressive conversions to a biventricular repair [15, 16]. Nathan and colleagues [17] report a series of 11 infants with DS and SVP who underwent biventricular conversion, with only 1 death but a high rate of reintervention. However, no hemodynamic data are reported, the median age at intervention is approximately 4 years, and median follow-up is only 18 months. Given the data in our study, it is hard to know whether the study by Nathan and colleagues [17] has a survival bias, in that healthier infants survived to undergo biventricular conversion. Given that survival in our study of infants with DS with SVP who have PVR less than 3 WUm2 is 100%, perhaps aggressive biventricular conversion should be reserved for patients with high PVR. Early management should be based on the patient’s expected surgical course. If suitability for a biventricular repair is in question, pulmonary blood flow must be regulated early in infancy. An appropriately tightened pulmonary artery band can protect the pulmonary vasculature in an unbalanced AVSD and does not prevent future biventricular or one-and-a-half ventricle repair. Pulmonary arterial pressure should also be considered when selecting the size of a systemic to pulmonary arterial shunt. Failure to protect the pulmonary vasculature early could jeopardize patient outcome if SVP is ultimately chosen. This study has multiple limitations. As with any retrospective review, data are limited by what was clinically
Table 3. Cox Regression Analysis Variable Univariable analysis Down syndrome, yes versus no PVR 3 WUm2 in first year, yes versus no Highest PVR in first year of life, per mm Hg Sex, male versus female Race Black versus white Hispanic versus white Other versus white Term versus preterm birth Systemic ventricle, left versus right AVVR pre-SCPC, > mild versus mild PVR pre-SCPC, per 1 WUm2 Mean PAP pre-SCPC, per mm Hg VEDP pre-SCPC, per mm Hg AVVR pre-TCPC, > mild versus mild PVR pre-TCPC, per 1 WUm2 Mean PAP pre-TCPC, per mm Hg Multivariable analysis Down syndrome, yes versus no PVR 3 WUm2 in first year, yes versus no
n
Died or Received a Transplant
Censored
58 30 30 58 58 . . . 42 49 54 30 30 29 32 30 30
17 7 7 17 17 . . . 9 15 15 7 6 6 3 2 2
41 23 23 41 41 . . . 33 34 39 23 24 23 29 28 28
30
7
23
HR (95% CI)
2.92 10.56 1.32 0.64
p Value
(1.03–8.31) (1.27–87.86) (1.08–1.63) (0.23–1.84) . (0.23–3.41) (0.53–5.02) (0.11–7.30) (0.14–2.19) (0.36–3.09) (0.85–7.35) (1.06–1.62) (0.99–1.09) (1.05–2.02) (0.20–24.60) (0.55–3.27) (0.75–3.06)
0.04 0.03 0.01 0.41 . 0.87 0.39 0.93 0.39 0.92 0.10 0.01 0.15 0.03 0.51 0.53 0.25
1.46 (0.27–7.75) 9.75 (1.14–83.45)
0.66 0.04
0.89 1.64 0.91 0.55 1.05 2.50 1.31 1.04 1.45 2.23 1.34 1.51
AVVR ¼ atrioventricular valve regurgitation; HR ¼ hazard ratio; PAP ¼ pulmonary artery pressure; pre-SCPC ¼ preoperative superior cavopulmonary connection; pre-TCPC ¼ preoperative total cavopulmonary connection; PVR ¼ pulmonary vascular resistance; VEDP ¼ ventricular end-diastolic pressure; WUm2 ¼ Wood unit m2.
Ann Thorac Surg 2016;-:-–-
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
7
Fig 3. Kaplan-Meier survival in patients (A) with and (B) without Down syndrome with single-ventricle anatomy who had pulmonary vascular resistance measured in the first year of life, stratified by pulmonary vascular resistance 3 WUm2. (PVR ¼ pulmonary vascular resistance; WUm2 ¼ Wood unit m2.)
collected. Cardiac catheterization was performed in only one-half of the patients. DS and SVP represent a rare combination of diseases, and unbalanced AVSD without heterotaxy is rare, which limits statistical power. Although we performed an exploratory multivariable analysis to examine the separate effects of DS and high PVR, interpretation of this analysis is limited, given such a small sample size. Medical and surgical management of patients with DS and functional single ventricle is patient specific. As such, there are inherent biases in the management strategy to achieve the optimal outcome for each patient, which created variability in timing of catheterization and timing/type of operation. Some patients currently on the single-ventricle pathway have not been completely ruled out for future attempts at biventricular repair or TCPC, and others, thought to be good candidates, may not be so. This study did not consider the presence of coexisting noncardiac structural abnormalities and comorbid conditions known to affect patients with DS. No other genetic conditions were reported in the medical record, but we cannot exclude the presence of unreported genetic abnormalities. Surgical eras were reported but not included in the analysis. Although our data are limited by the rarity of this population, we conclude that patients with DS undergoing SVP remain a high-risk group. However, the increased mortality in our study was associated with PVR 3 WUm2 or greater in the first year of life, and DS was not an independent risk factor for mortality when accounting for PVR. Overall, survival was excellent in patients with DS undergoing SVP with PVR less than 3 WUm2 in the first year of life, with minimal mortality even in those with PVR greater than 3 WUm2 beyond 2 years of age. Definitive palliation can be successful in appropriately
selected patients. Early, aggressive efforts to limit pulmonary arterial pressure can serve as a bridge to decision in borderline situations.
References 1. Pandit C, Fitzgerald DA. Respiratory problems in children with down syndrome. J Paediatr Child Health 2012;48: E147–52. 2. Campbell RM, Adatia I, Gow RM, Webb GD, Williams WG, Freedom RM. Total cavopulmonary anastomosis (Fontan) in children with Down’s syndrome. Ann Thorac Surg 1998;66: 523–6. 3. D’Alto M, Mahadevan VS. Pulmonary arterial hypertension associated with congenital heart disease. Eur Respir Rev 2012;21:328–37. 4. Evans JM, Dharmar M, Meierhenry E, Marcin JP, Raff GW. Association between Down syndrome and in-hospital death among children undergoing surgery for congenital heart disease: a US population-based study. Circ Cardiovasc Qual Outcomes 2014;7:445–52. 5. Fudge JC Jr, Li S, Jaggers J, et al. Congenital heart surgery outcomes in Down syndrome: analysis of a national clinical database. Pediatrics 2010;126:315–22. 6. Furukawa T, Park IS, Yoshikawa T, et al. Outcome of univentricular repair in patients with Down syndrome. J Thorac Cardiovasc Surg 2013;146:1349–52. 7. Gupta-Malhotra M, Larson VE, Rosengart RM, Guo H, Moller JH. Mortality after total cavopulmonary connection in children with the down syndrome. Am J Cardiol 2010;105:865–8. 8. Wada N, Takahashi Y, Ando M, Park IS, Sasaki T. Single ventricle repair in children with Down’s syndrome. Gen Thorac Cardiovasc Surg 2008;56:104–8. 9. Sakurai H, Akita T, Kato N, et al. Short-term results of the Fontan operation in patients with Down syndrome [in Japanese]. Kyobu Geka 2005;58:215–8. 10. Song J, Kang IS, Huh J, et al. Interstage mortality for functional single ventricle with heterotaxy syndrome: a retrospective study of the clinical experience of a single tertiary center. J Cardiothorac Surg 2013;8:93.
8
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
11. Chiu SN, Shao PL, Wang JK, et al. Severe bacterial infection in patients with heterotaxy syndrome. J Pediatr 2014;164:99–104.e1. 12. Alsoufi B, Manlhiot C, Awan A, et al. Current outcomes of the Glenn bidirectional cavopulmonary connection for single ventricle palliation. Eur J Cardiothorac Surg 2012;42:42–8; discussion 48–9. 13. 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;136:1237–42. 14. Hawkins A, Langton-Hewer S, Henderson J, Tulloh RM. Management of pulmonary hypertension in Down syndrome. Eur J Pediatr 2011;170:915–21.
Ann Thorac Surg 2016;-:-–-
15. Delmo Walter EM, Ewert P, Hetzer R, et al. Biventricular repair in children with complete atrioventricular septal defect and a small left ventricle. Eur J Cardiothorac Surg 2008;33:40–7. 16. Oshima Y, Maruo A, Hasegawa T, Matsuhisa H. Repair of unbalanced atrioventricular septal defect with small right ventricle [published ahead of print July 24, 2014]. Asian Cardiovasc Thorac Ann 2016;24:78–80. http://dx.doi.org/10. 1177/0218492314546029. 17. Nathan M, Liu H, Pigula FA, et al. Biventricular conversion after single-ventricle palliation in unbalanced atrioventricular canal defects. Ann Thorac Surg 2013;95:2086–95; discussion 2095–6.
Background. We describe survival in patients with Down syndrome (DS) with single-ventricle anatomy and palliation and characterize risk factors associated with mortality. Methods. All patients with DS and single-ventricle anatomy documented in the electronic medical record at our institution from January 1, 1992, to May 1, 2014, were compared with patients with unbalanced atrioventricular septal defects and single-ventricle anatomy, without DS or heterotaxy, during the same period. Survival analysis was performed to evaluate factors associated with mortality, including the presence of DS. Results. There were 28 patients with DS and 30 patients without DS. One-year survival with DS was 75% (95% CI: 55% to 87%); 5-year survival was 61% (95% CI: 40% to 76%). All DS deaths except one occurred before 2 years of age. One-year non-DS survival was 93% (95% CI: 76% to 98%); 5-year survival was 85% (95% CI: 64% to 94%). Factors associated with
death by univariable analysis included DS (p [ 0.04), pulmonary vascular resistance (PVR) of at least 3 Wood units 3 meter2 (WUm2) in the first year of life (p [ 0.03), and moderate-to-severe atrioventricular valve regurgitation (p [ 0.1). In combined analysis, when accounting for PVR of at least 3 WUm2 (hazard ratio [HR] 9.8, 95% CI: 1.1 to 83.5, p [ 0.04), DS was not associated with increased mortality (HR 1.5, 95% CI: 0.3 to 7.8, p [ 0.66). No patient with DS with PVR less than 3 WUm2 died. Conclusions. Children with DS and single-ventricle anatomy have excellent survival when PVR is less than 3 WUm2 in the first year of life, with minimal mortality beyond 2 years of age. When accounting for PVR, DS alone is not associated with increased mortality in patients with single-ventricle anatomy.
P
patients with and without DS who undergo SVP, to characterize risk factors associated with mortality in DS and SVP, and to investigate whether DS alone is a risk factor for death.
atients with Down syndrome (DS) and congenital heart disease who undergo single-ventricle palliation (SVP) represent a population considered to be at high risk for morbidity and mortality compared with their euploidic peers. Low transpulmonary pressures are necessary for successful palliation, but children with DS are inherently predisposed to higher pulmonary vascular resistance (PVR) [1–3]. Recent large administrative database studies have shown increased hospital deaths after staged palliation procedures in patients with DS compared with patients without DS [4, 5]. Other studies have reported on mortality in DS after total cavopulmonary connection (TCPC) [2, 6–9], with inconsistent findings about mortality risk compared with patients without DS. Overall, little has been published on the clinical and surgical outcomes of patients with DS, particularly before TCPC. The purpose of this study is to describe survival in patients with DS and single-ventricle anatomy, to compare survival in
Accepted for publication Nov 17, 2015. Address correspondence to Dr Kyle, Texas Children’s Hospital, Baylor College of Medicine, Department of Pediatrics, Division of Pediatric Cardiology, 6621 Fannin, MC# 19345-C, Houston, TX 77030; email: [email protected].
Ó 2016 by The Society of Thoracic Surgeons Published by Elsevier
(Ann Thorac Surg 2016;-:-–-) Ó 2016 by The Society of Thoracic Surgeons
Patients and Methods This study was approved by the Institutional Review Board at Baylor College of Medicine. Patients cared for between January 1, 1992, and May 1, 2014, were identified with institutional electronic databases. Search criteria included diagnosis of DS and single-ventricle anatomy. Single-ventricle anatomy was defined as the inability to surgically separate pulmonary and systemic cardiac output. For patients who underwent a surgical procedure, this included pulmonary artery banding, pulmonary artery banding with superior cavopulmonary connection (SCPC; also known as pulsatile Glenn), and intact pulmonary artery with SCPC (also known as one-and-a-half repair). In most patients, suitability for septation was determined by echocardiogram; however, in some, it was determined on intracardiac inspection during the operation. Patients without DS who had an unbalanced atrioventricular septal defect (AVSD) considered unsuitable for septation were selected for comparison with the 0003-4975/$36.00 http://dx.doi.org/10.1016/j.athoracsur.2015.11.047
2
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
use of the same search criteria. Patients with unbalanced AVSD were chosen for comparison, given that this lesion was present in most of the DS cohort. Patients with heterotaxy complex (isomerism) were excluded because of a known increased risk of death [10, 11]. Clinical diagnosis, demographic characteristics, gestational age, dominant ventricle (morphologic left versus right), age at surgical procedure, and type of pre-SCPC surgical procedure (pulmonary artery band, systemic-to-pulmonary artery shunt, or other) were recorded. Preterm was defined as birth less than 37 weeks’ gestation. Preoperative data collected included degree of atrioventricular valve regurgitation, mean pulmonary artery pressure, ventricular end-diastolic pressure, and calculated PVR as obtained by echocardiography and cardiac catheterization. The degree of atrioventricular valve regurgitation was estimated qualitatively and was divided into 1, none to mild, and 2, moderate to severe. Highest PVR in the first year of life was collected when available. PVR is reported in Wood unit meter2 (WUm2). Pre-SCPC PVR was defined as the PVR measured on the catheterization most closely preceding the SCPC surgical procedure.
Statistical Methods Wilcoxon rank sum and Pearson c2 tests were used to compare continuous and categorical variables, respectively, and Cochran-Armitage trend test was applied for ordinal variables. Cardiac survival (from birth to death or heart transplantation) was estimated with the KaplanMeier method. Birth was time zero, and all living patients without transplantation were censored at the date of last contact. Kaplan-Meier analysis was first used, including patients with and without DS, to perform univariable analysis to determine which factors were associated with mortality. A receiver operating characteristic curve analysis was created to identify a threshold value of PVR that maximized sensitivity and specificity for mortality to delineate a clinically useful cutoff. An exploratory multivariable Cox regression model was then created, initially including all covariates associated with mortality with p less than 0.2 by univariable analysis. Given the small dataset, backward elimination was used to discard covariates with p values greater than 0.05, retaining DS in the model to assess its effect estimate. Given high collinearity between PVR and mean pulmonary arterial pressure, only PVR was used in the model. A p value less than 0.05 was considered statistically significant. Analysis was performed with SAS version 9.4 (SAS Institute, Cary, NC).
Results Twenty-eight patients with DS and single-ventricle anatomy and 30 patients without DS with unbalanced AVSD were included. Patient characteristics are listed in Table 1. Median age at last follow-up was approximately 5 years in both groups (range, 2 months to 33 years in both). Statistically significant differences between the groups included older median age at SCPC (p ¼ 0.03), higher preSCPC PVR (p ¼ 0.02), and higher pre-SCPC mean
Ann Thorac Surg 2016;-:-–-
pulmonary artery pressure (p ¼ 0.01) in the DS group. The DS group was evenly divided between morphologic left (50%) and right (50%) ventricles, whereas 70% of the non-DS group had a morphologic right ventricle (p < 0.01). The majority of patients with DS had an unbalanced AVSD (26 of 28, 93%); one patient had hypoplastic left heart syndrome and another had tricuspid atresia. The clinical course of patients with DS is shown in Figure 1. The majority of patients were initially palliated with a pulmonary artery band (22 of 28, 79%), including 5 patients with a concomitant arch repair. Three had a Blalock-Taussig shunt (3 of 28, 11%), 1 of whom underwent a Norwood procedure. Of the remaining 3, 1 died before any operation and the other 2 underwent SCPC as their first palliation. SCPC was performed in 16 patients with DS (16 of 28, 57%) at a median age of 1.1 years (range, 3 months to 13.4 years). In contrast, median age at SCPC in patients without DS was 6 months (range, 3 months to 6.6 years; p ¼ 0.03). The median SCPC date for patients with DS and patients without DS was September 12, 2006 (range, June 23, 1997 to March 8, 2011) and March 18, 2008 (range, February 1, 1992 to April 14, 2014), respectively. Of the DS group, 8 remain with only SCPC, 3 have undergone TCPC, 1 has undergone a one-and-a-half ventricle repair, 2 died before another operation, and 2 underwent SCPC takedown for failing physiology. PVR before takedown in those 2 patients was 5.4 and 3.4. Data, including current management plan for patients with DS with SCPC, are found in Table 2. Presently, 3 patients who underwent SCPC are considered candidates for TCPC and are awaiting an operation. Two patients with DS underwent single-stage TCPC after pulmonary artery banding (ages at TCPC were 1.9 and 3.5 years). Thus, a total of 5 patients with DS achieved TCPC (18%; 5 of 28), at a median age of 3.5 years (range, 1.9 to 4.8 years). The median TCPC date for patients with DS and patients without DS was July 9, 2007 (range, June 23, 1997, to April 28, 2009) and September 19, 2009 (range, January 18, 1995, to June 17, 2013), respectively. The patients with DS with hypoplastic left heart syndrome and tricuspid atresia both successfully achieved TCPC. Four of the 5 are alive (age range at follow up, 5 to 11 years). TCPC was achieved in 16 patients without DS (53%). Sixteen patients with DS (57%) and 15 patients without DS (50%) underwent cardiac catheterization in the first year of life. Hemodynamic data are listed in Table 1. In patients with DS, 1-year Kaplan-Meier survival was 75% (95% CI: 55% to 87%); 5-year survival was 61% (95% CI: 40% to 76%). All deaths occurred before 2 years of age, except for 1 patient, which occurred at age 18 years. For patients without DS, 1-year Kaplan-Meier survival was 93% (95% CI: 76% to 98%); 5-year survival was 85% (95% CI: 64% to 94%; univariable comparison of overall Kaplan-Meier survival in DS and non-DS p ¼ 0.03). Median age at death was 11 months (range, 2 months to 18 years) for patients with DS and 2.3 years (range, 2 months to 5.7 years) for patients without DS. No patients with DS received a transplant. One patient without DS received a transplant; another is currently on the waiting list. Data
Ann Thorac Surg 2016;-:-–-
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
3
Table 1. Patient Characteristics Variable Male sex, n (%) Race/ethnicity, n (%) White Hispanic Black Other Lesion, n (%) Unbalanced AVSD HLHS Tricuspid atresia Systemic ventricle, n (%) Morphologic left Morphologic right Indeterminate Gestational age, n (%) Preterm Term No data Age at last follow-up, years, median (range) PVR 3 WUm2 in first year of life, n (%) Yes No No data Highest PVR in first year, WUm2, median (range) [n] First surgical procedure, n (%) PAB BT shunt SCPC Norwood palliation Other No data None Pre-SCPC AV valve regurgitation, n (%) None-mild Moderate-severe No data Age at pre-SCPC catheterization, years, median (range) [n] PVR pre-SCPC, WUm2, median (range) [n] Mean PAP pre-SCPC, mm Hg, median (range) [n] Age at SCPC, years, median (range) [n] LOS at SCPC, days, median (range) [n] Pre-TCPC AV valve regurgitation, n (%) None-mild Moderate-severe No TCPC PVR pre-TCPC, WUm2, median (range) [n] Mean PAP pre-TCPC, mm Hg, median (range) [n] Age at TCPC, years, median (range) [n] LOS at TCPC in days, median (range) [n] Age at death or heart transplantation, years, median (range) [n]
Non-Down Syndrome (n ¼ 30) 13 (43) 16 6 6 2
Down Syndrome (n ¼ 28) 7 (25)
(53) (20) (20) (7)
14 6 6 2
p Value 0.14
(50) (21) (21) (7)
1.00
30 (100) . .
26 (93) 1 (3) 1 (3)
0.33
2 (7) 21 (70) 7 (23)
13 (46) 13 (46) 2 (7)
<0.01
6 16 8 5.2
(20) (73) (27) (0.2–33.6)
4 16 8 5.1
(14) (57) (29) (0.2–33.1)
0.58
3 11 16 1.9
(10) (37) (53) (0.8–8.4) [14]
9 7 12 3.1
(32) (25) (43) (0.4–12.8) [16]
0.05
12 7 1 7 1 2
(40) (23) (3) (23) (3) (7)
22 2 2 1
(79) (7) (7) (3) . . 1 (3)
. 11 17 2 0.5 1.7 13 0.5 10
(37) (57) (7) (0.2–1.3) [15] (0.6–5.8) [15] (6–23.5) [15] (0.3–6.6) [26] (5–95) [22]
16 10 2 0.6 2.7 18 1.1 10
(57) (36) (7) (0.2–18.1) [16] (0.4–12.8) [15] (9–55) [15] (0.3–13.4) [19] (5–51) [16]
6 13 11 1.5 12 4.2 14 2.3
(20) (43) (37) (0.4–5.7) [18] (7–16) [18] (1.4–23.3) [16] (6–24) [16] (0.2–5.7) [5]
11 2 15 2.3 12.3 3.6 11 0.9
(39) (7) (54) (0.7–4.7) [12] (7–16) [12] (1.9–4.8) [5] (7–57) [5] (0.2–18) [11]
AV ¼ atrioventricular; AVSD ¼ atrioventricular septal defect; BT ¼ Blalock-Taussig; length of stay; PAB ¼ pulmonary artery band; PAP ¼ pulmonary artery pressure; superior cavopulmonary connection; TCPC ¼ total cavopulmonary connection.
HLHS ¼ hypoplastic left heart syndrome; PVR ¼ pulmonary vascular resistance;
0.3
0.08 0.02
0.10
0.55 0.02 0.01 0.03 0.63 <0.01
0.25 0.48 0.21 0.5 0.46 LOS ¼ SCPC ¼
4
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
Ann Thorac Surg 2016;-:-–-
Fig 1. Management and outcomes in children with Down syndrome and FSV. 1Included 1 Norwood arch reconstruction. 2One patient still has PAB; SCPC delayed because of PVR ¼ 10; group also includes five arch repairs. 3Attempted biventricular repair was unsuccessful and was taken down to PAB with subsequent demise. 4Takedown performed for failing SCPC physiology; 1 of 2 patients died. 5Developed progressive cyanosis/ exercise intolerance many years after PAB; could not septate, instead underwent palliative shunting. *Three patients who underwent SCPC are considered candidates for and await TCPC; if successful this percentage would increase to 32%. (BiV ¼ biventricular; BT ¼ Blalock-Taussig; FSV ¼ functional single ventricle; PAB ¼ pulmonary artery band; SCPC ¼ superior cavopulmonary connection; TCPC ¼ total cavopulmonary connection.)
related to deaths in the patients with DS can be found in Table 2. A receiver operating characteristic curve was created for PVR with the outcome of death or transplantation in all patients to determine a clinically useful cutoff for PVR, with the goal of maximizing sensitivity and specificity. A value of 2.8 WUm2 or more for PVR in the first year of life resulted, with a sensitivity of 86% and specificity of 74% (Fig 2). For clinical purposes, a value of 3 WUm2 was used (same sensitivity and specificity), and subsequent analyses that include PVR used values less than 3 WUm2 and 3 WUm2 or more. On univariable analysis, factors associated with death or transplantation included the presence of DS, PVR of at least 3 WUm2 in the first year of life, higher mean pulmonary arterial pressure before SCPC, worse atrioventricular valve regurgitation before SCPC, and higher ventricular end-diastolic pressure before SCPC (Table 3). No patients with DS and a PVR less than 3 WUm2 died (Fig 3). By univariable analysis, the hazard ratio (HR) for mortality for PVR of at least 3 WUm2 in the first year of life was 10.6 (95% CI: 1.3 to 87.9). By multivariable analysis (Table 3), the only variable that remained significantly associated with mortality was
PVR of at least 3 WUm2 in the first year of life (HR 9.75, 95% CI: 1.14 to 83.45). DS itself was not associated with increased mortality (HR 1.5, 95% CI: 0.3 to 7.8, p ¼0.66).
Comment To our knowledge, this is the largest single-center report of outcomes in patients with DS and single-ventricle anatomy. Our study confirms that SVP in patients with DS is associated with appreciable mortality; however, this appears to be accounted for by elevated PVR, because patients with DS and PVR less than 3 WUm2 had 100% survival in our study. PVR may be useful in risk stratifying this cohort to identify those at highest risk. Elevated PVR is a well-documented risk factor for poor outcomes with SVP [12, 13]. In a 2012 study of 227 patients with variable single-ventricle disorders, Alsoufi and colleagues [12] showed that risk factors for death before TCPC included a PVR more than 3 WUm2. Our data support this association in the DS population, with a HR of 9.8 when PVR was at least 3 WUm2 in the first year of life. As discussed in the Results section, elevated PVR (3 WUm2) was noted in patients who did not progress in
Ann Thorac Surg 2016;-:-–-
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
5
Table 2. Patients With Down Syndrome Age Current patients with SCPC 7y 7y 9y 5y 8y 5y 17 y 18 y
Most Recent PVR, WUm2
Comments
<1 2.6 <1 2.4 4.7 2.3
Considered a TCPC candidate Considered a TCPC candidate Considered a TCPC candidate Complicated post-SCPC course; TCPC candidacy in question Complicated post-SCPC course; TCPC candidacy in question Concerning pulmonary-to-systemic venous collateral; TCPC candidacy in question Parental choice to defer catheterization and operation Severe developmental delay, not ambulatory; not a TCPC candidate
No data 2.3 Most Recent PVR, WUm2
Comments
Current patients with definitive palliation 8y 11 y 9y
1.3 2.5 1.9
5y
2.3
5y
1.6
HLHS; nonfenestrated extracardiac conduit TCPC at 2.5 years Tricuspid atresia; fenestrated extracardiac TCPC at 4.3 years RV dominant AVSD; nonfenestrated extracardiac TCPC at 4.8 years LV dominant AVSD; fenestrated lateral tunnel TCPC at 3.5 years LV dominant AVSD; one-and-a-half ventricle repair at 4 years
Age
Age at Death Deaths 2 mo 2 mo 4 mo 5 mo 8 mo 10 mo 14 mo 19 mo 20 mo 24 mo 18 y Childhood
PVR, WUm2
Anatomy and Surgical History
Cause of Death
7.3 No data No data 12.8 No data 3.9 No data 3.5 3 3.1 7.2 No data
RV dominant AVSD, coarctation s/p PAB and arch repair RV dominant AVSD, type A interrupted arch s/p PAB, arch repair RV dominant AVSD, s/p failed attempt at 2V repair, PAB, ECMO RV dominant AVSD, none LV dominant AVSD, pulmonary atresia s/p BT shunt RV dominant AVSD, arch hypoplasia s/p PAB, arch repair; SCPC LV dominant AVSD s/p PAB; SCPC; ECMO LV dominant AVSD s/p PAB; SCPC; SCPC takedown to PAB RV dominant AVSD s/p PAB; SCPC; BT shunt LV dominant AVSD s/p PAB; single-stage TCPC LV dominant AVSD s/p PAB RV dominant AVSD, arch hypoplasia s/p PAB, arch repair
VT, cardiac arrest Heart failure Heart failure Heart failure Cardiac arrest Cardiac arrest Heart failure Heart failure Heart failure Heart failure No data No data
AVSD ¼ atrioventricular septal defect; BT ¼ Blalock-Taussig; ECMO ¼ extracorporeal membrane oxygenation; HLHS ¼ hypoplastic left heart syndrome; LV ¼ left ventricle; PAB ¼ pulmonary artery band; PVR ¼ pulmonary vascular resistance; RV ¼ right ventricle; SCPC ¼ superior cavopulmonary connection; s/p ¼ status post; TCPC ¼ total cavopulmonary connection; VT ¼ ventricular tachycardia; 2V ¼ biventricular.
their palliation and in patients who failed SCPC. Even in patients with DS and PVR 3 WUm2 or more, mortality beyond 2 years of age was minimal in our study. Previous literature has focused on TCPC outcomes in DS. In a 2010 study, Gupta-Malhotra and colleagues [7] reported an early mortality after TCPC of 35% and found DS to be an independent variable associated with a significantly higher risk of mortality in the early postoperative period. In contrast, in 2013, Furukawa and colleagues [6] concluded that the overall mortality rate of patients with DS undergoing TCPC was lower (12.5%), with no significant difference from that of children without DS.
Evident in this and other studies is that TCPC is infrequently attempted in patients with DS [4, 5, 7]. In a 2010 study, there were only 17 TCPC operations among 4,350 patients with DS in The Society of Thoracic Surgeons Congenital Heart Surgery Database [5]. In the referenced single-institution studies [2, 6, 8, 9], the reported TCPC numbers were 4, 8, 3, and 2, respectively. Including the one patient who underwent a one-and-ahalf ventricle repair, the number achieving what would be considered definitive palliation in our cohort was 6 (6 of 28, 21%). The group of incompletely palliated patients with elevated PVR presents a clinical dilemma (see Table 2 for comments about patients who underwent
6
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
Fig 2. Receiver operating characteristic curve for PVR in the first year of life and death/cardiac transplantation. A value of 3 WUm2 for PVR had a sensitivity of 86% and specificity of 74%. (AUC ¼ area under the curve; PVR ¼ pulmonary vascular resistance; WUm2 ¼ Wood unit m2.)
SCPC not offered TCPC). The addition of pulmonary vasodilators may be reasonable and was used in some of our patients. Surgical relief of upper airway obstruction might be entertained [14].
Ann Thorac Surg 2016;-:-–-
Given the poor outcomes reported in DS and SVP, there has been some recent consideration of aggressive conversions to a biventricular repair [15, 16]. Nathan and colleagues [17] report a series of 11 infants with DS and SVP who underwent biventricular conversion, with only 1 death but a high rate of reintervention. However, no hemodynamic data are reported, the median age at intervention is approximately 4 years, and median follow-up is only 18 months. Given the data in our study, it is hard to know whether the study by Nathan and colleagues [17] has a survival bias, in that healthier infants survived to undergo biventricular conversion. Given that survival in our study of infants with DS with SVP who have PVR less than 3 WUm2 is 100%, perhaps aggressive biventricular conversion should be reserved for patients with high PVR. Early management should be based on the patient’s expected surgical course. If suitability for a biventricular repair is in question, pulmonary blood flow must be regulated early in infancy. An appropriately tightened pulmonary artery band can protect the pulmonary vasculature in an unbalanced AVSD and does not prevent future biventricular or one-and-a-half ventricle repair. Pulmonary arterial pressure should also be considered when selecting the size of a systemic to pulmonary arterial shunt. Failure to protect the pulmonary vasculature early could jeopardize patient outcome if SVP is ultimately chosen. This study has multiple limitations. As with any retrospective review, data are limited by what was clinically
Table 3. Cox Regression Analysis Variable Univariable analysis Down syndrome, yes versus no PVR 3 WUm2 in first year, yes versus no Highest PVR in first year of life, per mm Hg Sex, male versus female Race Black versus white Hispanic versus white Other versus white Term versus preterm birth Systemic ventricle, left versus right AVVR pre-SCPC, > mild versus mild PVR pre-SCPC, per 1 WUm2 Mean PAP pre-SCPC, per mm Hg VEDP pre-SCPC, per mm Hg AVVR pre-TCPC, > mild versus mild PVR pre-TCPC, per 1 WUm2 Mean PAP pre-TCPC, per mm Hg Multivariable analysis Down syndrome, yes versus no PVR 3 WUm2 in first year, yes versus no
n
Died or Received a Transplant
Censored
58 30 30 58 58 . . . 42 49 54 30 30 29 32 30 30
17 7 7 17 17 . . . 9 15 15 7 6 6 3 2 2
41 23 23 41 41 . . . 33 34 39 23 24 23 29 28 28
30
7
23
HR (95% CI)
2.92 10.56 1.32 0.64
p Value
(1.03–8.31) (1.27–87.86) (1.08–1.63) (0.23–1.84) . (0.23–3.41) (0.53–5.02) (0.11–7.30) (0.14–2.19) (0.36–3.09) (0.85–7.35) (1.06–1.62) (0.99–1.09) (1.05–2.02) (0.20–24.60) (0.55–3.27) (0.75–3.06)
0.04 0.03 0.01 0.41 . 0.87 0.39 0.93 0.39 0.92 0.10 0.01 0.15 0.03 0.51 0.53 0.25
1.46 (0.27–7.75) 9.75 (1.14–83.45)
0.66 0.04
0.89 1.64 0.91 0.55 1.05 2.50 1.31 1.04 1.45 2.23 1.34 1.51
AVVR ¼ atrioventricular valve regurgitation; HR ¼ hazard ratio; PAP ¼ pulmonary artery pressure; pre-SCPC ¼ preoperative superior cavopulmonary connection; pre-TCPC ¼ preoperative total cavopulmonary connection; PVR ¼ pulmonary vascular resistance; VEDP ¼ ventricular end-diastolic pressure; WUm2 ¼ Wood unit m2.
Ann Thorac Surg 2016;-:-–-
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
7
Fig 3. Kaplan-Meier survival in patients (A) with and (B) without Down syndrome with single-ventricle anatomy who had pulmonary vascular resistance measured in the first year of life, stratified by pulmonary vascular resistance 3 WUm2. (PVR ¼ pulmonary vascular resistance; WUm2 ¼ Wood unit m2.)
collected. Cardiac catheterization was performed in only one-half of the patients. DS and SVP represent a rare combination of diseases, and unbalanced AVSD without heterotaxy is rare, which limits statistical power. Although we performed an exploratory multivariable analysis to examine the separate effects of DS and high PVR, interpretation of this analysis is limited, given such a small sample size. Medical and surgical management of patients with DS and functional single ventricle is patient specific. As such, there are inherent biases in the management strategy to achieve the optimal outcome for each patient, which created variability in timing of catheterization and timing/type of operation. Some patients currently on the single-ventricle pathway have not been completely ruled out for future attempts at biventricular repair or TCPC, and others, thought to be good candidates, may not be so. This study did not consider the presence of coexisting noncardiac structural abnormalities and comorbid conditions known to affect patients with DS. No other genetic conditions were reported in the medical record, but we cannot exclude the presence of unreported genetic abnormalities. Surgical eras were reported but not included in the analysis. Although our data are limited by the rarity of this population, we conclude that patients with DS undergoing SVP remain a high-risk group. However, the increased mortality in our study was associated with PVR 3 WUm2 or greater in the first year of life, and DS was not an independent risk factor for mortality when accounting for PVR. Overall, survival was excellent in patients with DS undergoing SVP with PVR less than 3 WUm2 in the first year of life, with minimal mortality even in those with PVR greater than 3 WUm2 beyond 2 years of age. Definitive palliation can be successful in appropriately
selected patients. Early, aggressive efforts to limit pulmonary arterial pressure can serve as a bridge to decision in borderline situations.
References 1. Pandit C, Fitzgerald DA. Respiratory problems in children with down syndrome. J Paediatr Child Health 2012;48: E147–52. 2. Campbell RM, Adatia I, Gow RM, Webb GD, Williams WG, Freedom RM. Total cavopulmonary anastomosis (Fontan) in children with Down’s syndrome. Ann Thorac Surg 1998;66: 523–6. 3. D’Alto M, Mahadevan VS. Pulmonary arterial hypertension associated with congenital heart disease. Eur Respir Rev 2012;21:328–37. 4. Evans JM, Dharmar M, Meierhenry E, Marcin JP, Raff GW. Association between Down syndrome and in-hospital death among children undergoing surgery for congenital heart disease: a US population-based study. Circ Cardiovasc Qual Outcomes 2014;7:445–52. 5. Fudge JC Jr, Li S, Jaggers J, et al. Congenital heart surgery outcomes in Down syndrome: analysis of a national clinical database. Pediatrics 2010;126:315–22. 6. Furukawa T, Park IS, Yoshikawa T, et al. Outcome of univentricular repair in patients with Down syndrome. J Thorac Cardiovasc Surg 2013;146:1349–52. 7. Gupta-Malhotra M, Larson VE, Rosengart RM, Guo H, Moller JH. Mortality after total cavopulmonary connection in children with the down syndrome. Am J Cardiol 2010;105:865–8. 8. Wada N, Takahashi Y, Ando M, Park IS, Sasaki T. Single ventricle repair in children with Down’s syndrome. Gen Thorac Cardiovasc Surg 2008;56:104–8. 9. Sakurai H, Akita T, Kato N, et al. Short-term results of the Fontan operation in patients with Down syndrome [in Japanese]. Kyobu Geka 2005;58:215–8. 10. Song J, Kang IS, Huh J, et al. Interstage mortality for functional single ventricle with heterotaxy syndrome: a retrospective study of the clinical experience of a single tertiary center. J Cardiothorac Surg 2013;8:93.
8
COLQUITT ET AL SURVIVAL IN DOWN SYNDROME AND SINGLE-VENTRICLE PALLIATION
11. Chiu SN, Shao PL, Wang JK, et al. Severe bacterial infection in patients with heterotaxy syndrome. J Pediatr 2014;164:99–104.e1. 12. Alsoufi B, Manlhiot C, Awan A, et al. Current outcomes of the Glenn bidirectional cavopulmonary connection for single ventricle palliation. Eur J Cardiothorac Surg 2012;42:42–8; discussion 48–9. 13. 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;136:1237–42. 14. Hawkins A, Langton-Hewer S, Henderson J, Tulloh RM. Management of pulmonary hypertension in Down syndrome. Eur J Pediatr 2011;170:915–21.
Ann Thorac Surg 2016;-:-–-
15. Delmo Walter EM, Ewert P, Hetzer R, et al. Biventricular repair in children with complete atrioventricular septal defect and a small left ventricle. Eur J Cardiothorac Surg 2008;33:40–7. 16. Oshima Y, Maruo A, Hasegawa T, Matsuhisa H. Repair of unbalanced atrioventricular septal defect with small right ventricle [published ahead of print July 24, 2014]. Asian Cardiovasc Thorac Ann 2016;24:78–80. http://dx.doi.org/10. 1177/0218492314546029. 17. Nathan M, Liu H, Pigula FA, et al. Biventricular conversion after single-ventricle palliation in unbalanced atrioventricular canal defects. Ann Thorac Surg 2013;95:2086–95; discussion 2095–6.