|
|
||||||||
a Division of Cardiology, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
b Division of General and Community Pediatrics, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
c Division of Cardiothoracic Surgery, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
d Division of Critical Care Medicine, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio
Accepted for publication January 12, 2011.
* Address correspondence to Dr Anderson, Cincinnati Children's Hospital Medical Center, The Heart Institute, ML2003, Cincinnati, OH 45255 (Email: jeffrey.anderson{at}cchmc.org).
| Abstract |
|---|
|
|
|---|
Methods: This retrospective case series examined children who underwent Fontan at our institution between January 2003 and December 2008. Weight and height were obtained at the time of admission for Fontan. Data from preoperative echocardiogram and cardiac catheterization were abstracted to document cardiac function and hemodynamic measurements. Outcome variables included ventilator time, chest tube duration, postoperative infections (bacteremia, mediastinitis, urinary tract infection, gastroenteritis, or culture-positive pneumonia), and length of hospital stay.
Results: Fifty-five patients were included for analysis. The median age at Fontan was 46 months (range, 18 to 72); median WAZ was –1.0 (–3.8 to +2.0), and height for age z-score was –1.1 (–3.7 to +1.5). The WAZ was less than –2.0 in 19% of patients. Multivariable modeling revealed that patients with a WAZ less than –2.0 (p = 0.006) had a greater incidence of serious postoperative infections. The only factor predicting longer length of hospital stay was presence of a serious postoperative infection (p < 0.0001). Ventilator time was predicted only by length of cardiopulmonary bypass (p = 0.01). No factors were associated with longer chest tube duration.
Conclusions: Growth failure in children with a single ventricle persists through presentation for Fontan. A WAZ less than –2.0 at Fontan is associated with a higher rate of serious postoperative infections, which are associated with longer length of hospital stay.
| Introduction |
|---|
|
|
|---|
The etiology of poor growth and malnutrition is poorly understood in the single ventricle population. It is known that these infants have a lower than desired total caloric intake and that their growth improves with the introduction of high-caloric enteral feeds [6]. It has also been shown that infants with a single ventricle have oral-motor coordination problems and are slower to reach goal caloric intake [12]. Finally, energy expenditure is increased during the initial hours after first-stage palliation in infants with a single ventricle [13], but that has not been evaluated thereafter.
Poor nutrition in adults is associated with longer hospital stays and increased rate of readmission [14, 15]. Among pediatric patients admitted to the hospital with serious infections, poor nutrition has been linked to increased morbidities [16]. Few studies have looked at the relationship between nutritional status and surgical outcomes in children undergoing cardiac surgery. Leite and associates [17, 18] found increased rates of mortality and postsurgical infection in a group of patients with diverse types of congenital heart disease, with poor nutrition defined as low serum albumin level or low weight-for-age percentile. We have previously shown an association between lower WAZ and poor short-term surgical outcomes after the bidirectional Glenn (BDG) procedure [19]. There are no prior studies, to our knowledge, relating nutritional status and weight gain to perioperative morbidity in patients undergoing the Fontan procedure.
The purpose of this study was to identify specific anthropometric and cardiovascular variables that predict lower WAZ at Fontan, and poorer short-term surgical outcomes and perioperative morbidity in infants with a single ventricle after the Fontan procedure.
| Patients and Methods |
|---|
|
|
|---|
Data Collection
Study data were collected from hospital medical records and Heart Institute outpatient records, as well as cardiac catheterization, echocardiography, and cardiac surgical databases. Data collected included demographic information including sex, gestational age and race; anthropometric measurements and information regarding nutritional management; pre-Fontan medical and surgical management including length of neonatal and post-BDG hospitalizations, documentation of vocal cord paralysis after neonatal surgery and the need for readmission between BDG and Fontan; echocardiographic and hemodynamic data; and data from the Fontan operative and perioperative periods.
Anthropometric measurements and nutritional management
Anthropometric measures included absolute weight and WAZ at initial neonatal hospital admission and discharge and at admission for BDG, as well as weight, length, WAZ, and height-for-age z-score at the time of the Fontan procedure. The WAZ and height-for-age z-scores were calculated using the Centers for Disease Control Epi-Info program (Centers for Disease Control, Atlanta, GA). In addition to absolute WAZ, we identified patients who met criteria for failure to thrive at the time of their Fontan who had a WAZ below –2.0.
Pre-BDG medical and surgical management
Information regarding medical and surgical management before the Fontan procedure was recorded and included initial cardiac diagnosis, previous surgeries and hospital courses, history of necrotizing enterocolitis or vocal cord paralysis, and outpatient medications at the time of the Fontan procedure.
Echocardiographic and hemodynamic data
Findings from the most recent echocardiogram performed before the Fontan procedure were reviewed. Atrioventricular valve regurgitation and aortic or neoaortic regurgitation were considered significant if graded as moderate or greater. Significant residual defects including restrictive atrial communication (>2 mm Hg gradient) and aortic arch obstruction (>20 mm Hg peak corrected gradient) were noted. Depressed ventricular function was defined as a qualitative description of moderate or greater systolic dysfunction in a right or left single ventricle or a shortening fraction less than 28% in any left ventricle. Data from the most recent cardiac catheterization performed before the Fontan procedure were reviewed. Measurements recorded included mean pulmonary artery pressure, transpulmonary gradient, pulmonary vascular resistance, ventricular end-diastolic pressure, aortic arch systolic pressure gradient, systemic arterial and mixed venous oxygen saturation, and hemoglobin level.
Operative data
Operative variables recorded included intraoperative mortality, cardiopulmonary bypass time, aortic cross-clamp time, circulatory arrest time, and concomitant procedures performed in addition to the Fontan.
Perioperative data
Variables recorded from the postoperative course included postoperative mortality (<30 days of surgery); total ventilator time (time from operating room to final extubation); chest tube duration (time from operating room to final chest tube removal); documented significant postoperative infections including bacteremia, mediastinitis, culture-positive urinary tract infections, culture-positive gastroenteritis, and pneumonia, which was defined as culture-positive endotracheal tube aspirate with chest radiograph changes consistent with pneumonia; significant pleural effusions that required change in diuretic regimen or insertion of new chest tube; episodes of reintubation; total intensive care unit days; and total hospital days.
Statistical Analyses
The primary outcome variables assessed were WAZ at Fontan and total hospital length of stay. Secondary outcome variables included the need for Fontan takedown, ventilator time, chest tube duration, significant infections, significant pleural effusions, and episodes of reintubation after Fontan. The predictor variables for WAZ at Fontan were WAZ at the time of BDG discharge, longer neonatal and post-BDG hospitalizations, and the need for readmission to the hospital between BDG and Fontan.
The primary predictor variable for total hospital length of stay and secondary outcome variables was WAZ at Fontan. Potential covariates for all outcome variables included diagnosis of hypoplastic left heart syndrome or variants, preoperative significant atrioventricular valve regurgitation, and aortic or neoaortic regurgitation or aortic arch obstruction, restrictive atrial communication, depressed ventricular function, and elevated mean pulmonary artery pressure or pulmonary vascular resistance. Operative covariate predictor variables include the length of cardiopulmonary bypass and aortic cross-clamp time after Fontan.
Statistical analysis was performed using Stata 10.0 analysis software (Stata Corporation, College Station, TX). All continuous variables were described and tested for central tendency to determine the normality of the data distribution. Normally distributed continuous variables are expressed as mean plus or minus standard deviation. Nonnormally distributed data are expressed using the median value and range. Potential dichotomous predictor variables that did not occur in more than 10% of our study population were not statistically evaluated. Bivariate analyses of potential risk factors for adverse outcome were performed using the Wilcoxon rank sum test for continuous variables and Fisher's exact test for dichotomous variables. Candidate predictor variables for multivariable modeling had a p value of less than 0.1 on bivariate analyses. Multivariable analyses were performed using linear regression. A correlation table was utilized to assess for colinearity of variables. A p value of less than 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
|
|
|
Information was available on feeding plans for 51 patients (93%) at neonatal discharge, 50 patients (91%) at BDG, and 53 patients (96%) at Fontan presentation. The number of patients exclusively fed orally steadily rose from 18 (29%) to 40 (71%) to 48 (87%), while the number of patients supported with either nasogastric or gastrostomy tube feedings fell from 40 (71%) to 18 (29%) to 9 (13%). There were 7 patients who had placement of gastrostomy tube: 2 during their neonatal hospitalization, 2 during the period before BDG, and 3 between BDG and Fontan. At the time of Fontan, there were 3 patients still requiring supplemental feeding utilizing their gastrostomy tube and 4 patients who were receiving overnight feeds through a nasogastric tube. Feeding plans were not standardized and were left to the discretion of individual cardiologists and pediatricians.
Echocardiographic and Hemodynamic Data
All patients had echocardiograms, and 54 of 55 patients underwent cardiac catheterization before surgery. Echocardiographic and hemodynamic findings are shown in Table 3.
|
Outcome Data
There was no postoperative mortality after Fontan in this group of patients. Analysis of the primary postoperative outcome variable demonstrated a median hospital length of stay of 8 days (5 to 51), and the median intensive care unit length of stay was 4 days (1 to 15). Analysis of secondary outcome variables demonstrated a median ventilator time of 9.5 hours (0 to 233), and a median chest tube duration of 4.5 days (2 to 14). Three patients required replacement of chest tubes after removal. Nine patients (17%) had serious postoperative infection including pneumonia (n = 4), mediastinitis (n = 2), urinary tract infection (n = 1), bacteremia (n = 1), and gastroenteritis (n = 1). There were no patients with multiple infections.
Multivariable Modeling
Complete results of analyses are seen in Table 4. Multivariable modeling revealed that the only factor associated with a WAZ at the time of Fontan was a WAZ at time of BDG discharge (p < 0.0001). No echocardiographic or hemodynamic factors were associated with a WAZ at Fontan. There was no association between underlying cardiac diagnosis of hypoplastic left heart syndrome or variant and a WAZ at Fontan.
|
| Comment |
|---|
|
|
|---|
Poor growth has been noted in infants with a single ventricle [10], and specifically in infants with hypoplastic left heart syndrome [6, 12, 20]. Infants with a single ventricle who have grown poorly are at risk for worse short-term outcome after the BDG procedure [19]. Our study has assessed the association between poor growth and short-term outcomes after the Fontan procedure. While there was some catch-up growth noted between the BDG procedure in these children, they tended to still be small for age in both weight and height at the time of their Fontan. Several studies have shown an improvement in growth after BDG palliation, but there have been mixed reports regarding the level of catch-up growth that these children experience [11, 21]. Vogt and coworkers [10] observed that there was a significant increase in z-scores for weight in the pre-Fontan period, which leveled off after completion of the Fontan. However, z-scores for height remained below normal both before and after Fontan completion. Cohen and associates [11] reported that children with single ventricles were small in both height and weight at presentation for Fontan and remained significantly underweight and shorter than the general population and their siblings in follow-up at 4.4 years (range, 3 to 5.8) after surgery. Although etiologies have been proposed for growth disturbances in children with a single ventricle, the mechanisms of this problem remain poorly understood and require further investigation.
Undernutrition and poor growth have an adverse effect on short-term postoperative outcomes in this population. The children entering Fontan with a WAZ less than –2.0 trended toward a longer hospital stay (p = 0.06). The only factor that was significantly associated with longer hospital length of stay was the presence of a serious infection in the postoperative period. We hypothesize that with a larger sample size the association between WAZ and hospital length of stay would become more evident. There was a trend toward an association between WAZ at the time of Fontan and the development of serious postoperative infections. The importance of this relationship became more apparent when the children with failure to thrive were evaluated separately. There was significant association between entering the Fontan procedure with a low WAZ (below –2) and the development of postoperative infections, indicating a threshold effect. The children with the most growth failure were at most risk for development of serious infections.
The relationship between poor nutrition and development of serious infections is well established [22]. Inadequate dietary intake leads to impaired growth, lowered immunity, mucosal damage, and susceptibility to invasion by pathogens [22]. Adequate preoperative nutrition is especially important in the postoperative period as nutritional deficiencies are known to impede wound healing [23]. De Mello and colleagues [24] demonstrated that children with poor nutritional status, defined as a WAZ less than –2, admitted to the intensive care unit for medical and surgical problems were more likely to develop culture-positive bloodstream infections. Poor nutrition in children undergoing cardiac surgery for a diverse number of congenital heart defects has also been associated with increase in postoperative infections [17]. Infants with a single ventricle undergo multiple palliative procedures during the first several years of life. Each time they have a surgical procedure, their nutritional status plays a role in their recovery.
Although there are no data on specific nutritional interventions that will improve growth in this high-risk population, there may be simple interventions that could improve their nutrition and growth. The addition of a nutrition specialist to team of individuals caring for infants with other forms of congenital heart disease has been shown to improve growth in the first few months of life [25]. Close monitoring by parents of weight gain with home scales has also been shown to lead to improved growth in patients with complex congenital heart disease [26].
Children with a single ventricle experience growth failure at a high rate. This growth failure is persistent through their presentation for the Fontan procedure. As we continue to improve surgical techniques that will improve mortality in this high-risk population, we must also look for ways to improve their nutrition and growth. Not only will these improvements lead to better short-term surgical outcome, but also improvement in nutrition may improve neurodevelopmental outcome, which is known to be poor in these patients.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. F. Cnota, K. R. Allen, S. Colan, W. Covitz, E. M. Graham, D. A. Hehir, J. C. Levine, R. Margossian, B. W. McCrindle, L. L. Minich, et al. Superior cavopulmonary anastomosis timing and outcomes in infants with single ventricle J. Thorac. Cardiovasc. Surg., May 1, 2013; 145(5): 1288 - 1296. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |