|
|
||||||||
Ann Thorac Surg 2003;76:1450-1456
© 2003 The Society of Thoracic Surgeons
a Division of Cardiovascular-Thoracic Surgery, Chicago, Illinois, USA
b Department of Speech and Language Pathology, Chicago, Illinois, USA
c Division of Gastroenterology, Hepatology, and Nutrition, Chicago, Illinois, USA
d Division of Cardiology, Children's Memorial Hospital, Chicago, Illinois, USA
Accepted for publication May 29, 2003.
* Address reprint requests to Dr Mavroudis, Division of Cardiovascular-Thoracic Surgery, Children's Memorial Hospital, 2300 Children's Plaza, M/C 22, Chicago, IL 60614, USA.
e-mail: cmavroudis{at}childrensmemorial.org
| Abstract |
|---|
|
|
|---|
METHODS: Fifty patients were evaluated after open heart operations with transesophageal echocardiography between March 1, 1999, and September 30, 1999. The diagnosis of dysphagia was made by a speech pathologist using a clinical swallowing evaluation. Potential predictors examined included demographic variables, anatomical diagnosis, surgical procedure, size of the transesophageal echocardiographic probe in relation to body size, length of probe insertion time, preoperative patient acuity status, duration of intubation, and time until discharge.
RESULTS: Dysphagia was found in 9 (18%) of the 50 patients. Risk factors identified were age of less than 3 years (odds ratio, 20.4; 95% confidence interval, 2.7 to 157; p = 0.002), intubation prior to operation (odds ratio, 17.7; 95% confidence interval, 9.4 to 210; p = 0.004), intubation for more than 7 days (odds ratio, 74.7; 95% confidence interval, 13.8 to 405; p = 0.001), and operation for left-sided obstructive lesions (odds ratio, 1.9; 95% confidence interval, 2.2 to 8.3; p = 0.038). The size of the transesophageal echocardiographic probe in relation to the weight of the patient was found to be predictive (p = 0.0001) of dysphagia. Vocal cord paralysis was noted in 4 (8%) of the 50 patients postoperatively. Adverse events related to aspiration occurred in 2 patients (4%). At discharge, nasogastric tube feedings were required in 6 patients (12%), and thickened feedings were recommended for 3 (6%) of the 50 patients. Resolution of dysphagia ranged from 13 to 150 days.
CONCLUSIONS: Eighteen percent of patients had dysphagia after an open heart operation with transesophageal echocardiography. Age of less than 3 years, preoperative patient acuity status, longer intubation times, and operation for left-sided obstructions are risk factors for dysphagia in this cohort of pediatric patients. The size of the transesophageal echocardiography probe in relation to the patient's weight was predictive of dysphagia. Physicians should consider using the new mini-multiplane transesophageal echocardiographic probes in patients weighing less than 5.5 kg. Vigilance in monitoring for the signs of preoperative and postoperative dysphagia with prompt referral to a speech therapist can substantially reduce patient morbidity, length of hospital stay, and requirement of prolonged nasogastric tube use.
| Introduction |
|---|
|
|
|---|
Dysphagia is an alteration in one or more phases of the swallowing process [1]. For the purposes of this study, we defined dysphagia or swallowing dysfunction as any alteration in the oral, pharyngeal, or esophageal phases of the swallowing process. Dysphagia is associated with increased morbidity and potential mortality. The incidence of dysphagia after open heart operations in adults is about 3% to 4%, but the incidence in children and adolescents is unknown [2, 3]. The purpose of this prospective study of postoperative dysphagia was to determine its incidence, predictors, risk factors, and outcomes in children and adolescents who have undergone an open heart operation using intraoperative transesophageal echocardiography (TEE).
| Material and methods |
|---|
|
|
|---|
Fifty consecutive patients ranging in age from newborn to 17 years and scheduled to undergo an open heart procedure with the use of intraoperative TEE between March 1, 1999, and September 30, 1999, were enrolled in the study. Exclusion criteria included trisomy 21, gastrointestinal disorders, history of direct aspiration or vocal cord paralysis, oral motor dysfunction, airway anomalies, cleft lip or palate, neurological disorders (greater than grade II intraventricular hemorrhage, stroke, intractable seizures, cerebral palsy, severe congenital anoxia), and prematurity (gestation age < 35 weeks). Patient characteristics of the entire cohort are described in Table 1.
Seven (43.8%) of 16 patients in the infant subgroup were neonates (
3 months of age), and 4 (57%) of the 7 were less than 1 week old.
|
|
The lesions in the study population are representative of the types of lesions encountered at our institution. However, during open enrollment for this study, there was an unusually high number of patients with complex single-ventricle morphology, age younger than 3 years, and higher acuity as determined by preadmission status, including the 2 patients who were waiting for cardiac transplantation.
Intraoperative variables
Procedure-related variables included type of surgical repair, ease and number of transesophageal echocardiographic (TEE) intubation attempts, length of TEE probe insertion time, size of TEE probe in relation to the weight of the patient, and cardiopulmonary bypass time. Sixteen different types of surgical repairs were performed. Twelve (24%) of the 50 participants underwent ventricular septal defect closure, 6 (12%) underwent tetralogy of Fallot repair, 6 (12%) had a Fontan procedure, and 3 (6%) underwent a stage-one Norwood operation.
All patients underwent operation with the use of TEE. At the time of this study, the Division of Cardiovascular-Thoracic Surgery at Children's Memorial Hospital routinely used intraoperative TEE during open heart surgical procedures as a valuable tool for confirmation of complex cardiac diagnosis, for confirmation of optimal surgical results, or both. Depending on the weight of the patient, one of three sizes of Hewlett-Packard TEE probes (Andover, MA) was used: the pediatric TEE biplane probe, recommended for patients weighing less than 20 kg, in 33 patients (66%); the adult biplane TEE probe, recommended for patients weighing 20 kg or more, in 3 (6%); and the omniplane TEE probe, recommended for patients weighing more than 25 kg, in 14 (28%) [5]. Length of probe insertion time ranged from 50 to 388 minutes (mean time, 162.3 minutes; median time, 149.5 minutes; standard deviation, 68.4 minutes). The person responsible for inserting the TEE probe varied. The surgeon advanced the probe in 12 patients (24%); the cardiologist, 36 (72%); and the anesthesiologist, in 2 (4%). Advancement of the probe was classified as difficult in 4 (8%) of the 50 patients, and 3 of them required a second attempt at probe placement.
Postoperative variables
Duration of hospital stay (days), length of time spent in the intensive care unit (days), and time until extubation (days) were all recorded. Symptoms experienced after extubation were also documented. All patients underwent a speech therapy evaluation coinciding with the advancement to oral or nasogastric bolus tube feedings. Depending on the clinical presentation, patients were evaluated a total of two to three times by a speech therapist. Patients were observed during meals to determine the tolerance of oral feedings, the ability to meet volume or dietary goals, the presence of coughing or congestion during meals, and the association of vomiting, desaturation, or increased work of breathing with meals. On the basis of the speech therapist's findings, one or all of the following were prescribed: a videofluoroscopic swallowing study to assess oropharyngeal motility of various food consistencies and to rule out direct aspiration; an evaluation by an otolaryngologist for the presence of airway anomalies, vocal cord paralysis, or both; and an alternative feeding regimen if aspiration was documented on the videofluoroscopic swallowing study. Chest radiographs were monitored for the presence of right middle lobe infiltrates.
Follow-up data were collected 1 week after discharge during the postoperative clinic visit and 1 month after discharge through a telephone interview. Information regarding the use of alternative feeding regimens, GER medications, and the ability to gain weight was obtained on all patients. Patients requiring long-term nasogastric tube use were followed until resolution of swallowing dysfunction occurred.
Statistical analysis
The major outcome variable was the development of dysphagia as diagnosed by a speech therapist using a clinical swallowing evaluation. Predictors of swallowing dysfunction were determined by means of univariate and multivariate analysis. For univariate data analysis, the Mann-Whitney test, Kruskal-Wallis analysis of variance, and the Fisher exact test were used, and a p value of 0.05 was designated as the criterion for variable inclusion in the multivariate analysis model. Multivariate analysis was conducted using a logistic regression model. Results are presented as odds ratios with 95% confidence intervals.
Because our institution uses intraoperative TEE in all patients undergoing open heart procedures, the direct effect of TEE on the development of swallowing dysfunction could not be evaluated. However, variables associated with TEE use, such as ease of TEE probe insertion, person responsible for TEE probe placement, duration of TEE probe insertion, and size of TEE probe compared with weight of the patient, were analyzed. Given that the recommendations for pediatric TEE probe application are based on the patient's body weight, an effort was made to determine the cutoff weight predictive of dysphagia. Optimal Data Analysis (Chicago, IL) was used to identify the numeric value among the 42 patients in whom the pediatric biplane probe was used that distinguished the 9 patients who developed dysphagia from the 33 patients who did not. The value of the cutoff weight was validated by a leave-one-out procedure in which the weight of each patient was used to predict dysphagia using a cutoff value based on data from the other 32 patients. The classification results for each of the patients were then merged to reveal the overall unbiased percentage accuracy.
| Results |
|---|
|
|
|---|
|
|
Length of intubation time ranged from day 0 (extubated in the operating room) to postoperative day 17 (mean time, 2.5 days; median time, 1 day; standard deviation, 3.4 days). Two patients required reintubation at 1 day and 2 days after the initial attempt at extubation. The number of postoperative days in the intensive care unit ranged from 1 to 19 (median time, 3 days; mean time, 4.5 days; standard deviation, 4.2 days). Time until initiation of full oral feeding ranged from 0 to 22 days (median time, 1 day; mean time, 2.3 days; with standard deviation, 2.6 days). Length of time until full oral feedings were tolerated was significantly affected by whether the patient was taking oral feedings before operation (p = 0.0002). Nineteen (38%) of the patients required nasogastric tube feedings. The time requirement ranged from 1 day after operation until many months after discharge. Nasogastric tube use correlated significantly with a negative history of preoperative oral feedings (p = 0.0001). Of the 19 patients who required nasogastric tube feedings, 9 (47%) were discharged home on a modified feeding regimen, either supplemental nasogastric tube feedings or thickened oral feedings. Length of hospital stay for the entire study group ranged from 2 to 25 postoperative days (mean time, 8.2 days; standard deviation, 5.7 days).
A number of clinical symptoms were identified as markers for dysphagia. The presence of specific symptoms after extubation was also found to be predictive of swallowing dysfunction. The sensitivity and the specificity of these symptoms as predictors of dysphagia are shown in Figure 1. Stridor and an altered cry or voice after extubation were noted in 3 (33.3%) of the 9 patients with swallowing dysfunction. Feeding intolerance as determined by the presence of a cough during feedings, vomiting, and symptoms of GER were noted in 4 (8%) of the 50 study participants. Increased work of breathing during feedings was noted in 7 (14%) patients, and poor oral feedings in the form of decreased volume was noted in 9 (18%). None of the patients experienced desaturation during feedings. Of note, 10 additional patients, all newborns, had an uncoordinated sucking-swallowing breathing pattern. Similar to the patients with dysphagia, the infants with uncoordinated oral motor patterns tended to possess a negative history of preoperative oral feedings (p = 0.001) and required supplemental nasogastric tube feedings after operation. All patients with dysphagia were discharged home on alternative feeding regimens.
|
|
| Comment |
|---|
|
|
|---|
Many pediatric studies [3, 4, 7] have described the complications associated with TEE insertion including airway obstruction, common pulmonary vein compression, vascular compression, tracheal extubation, esophageal perforation, gastric perforation, and dental injury. The results of this study identified TEE probe size in relation to patient weight as a risk factor for dysphagia. However, unlike adult studies, our results showed no other factors associated with TEE to be predictive of swallowing dysfunction. The reason for this may be that the lack of a comparison group and the small sample size caused the numbers in each of the subgroups within each TEE variable under investigation to be too small to reach significance.
This study did demonstrate a substantially higher incidence of swallowing dysfunction than reported in previous studies with adults who underwent an open heart operation [2]. One reason for the increased incidence found in our study may be the percentage of infants and children less than 3 years of age (54%). Differences in anatomical size and physical relationship of the oral structures, especially the relative size of the infant's tongue in proportion to the oral cavity and the higher and more anterior position of the larynx, may place the child at higher risk for dysphagia [12]. These anatomical differences provide the rationale for all of the patients in our study displaying dysphagia in the form of oropharyngeal dysfunction in contrast to their adult counterparts, who experienced impairment in all four phases of the swallowing process [13, 14]. The other factor that contributed to our higher incidence of dysphagia was the percentage of patients with complex cardiac lesions (42%). This variable in conjunction with younger age most likely represents the subgroup of patients who are more acutely ill, require prolonged intubation, and are at higher risk for vocal cord paralysis.
We observed in our cohort that the presence of dysphagia adversely affected postoperative recovery and contributed to major morbidity. The etiology of these complications reflects the infant's or young child's difficulty in effectively protecting or clearing the airway [15]. Several studies have documented the strong association between dysphagia and unilateral vocal cord paralysis. Bhattacharyya and colleagues [16] studied 64 patients with unilateral vocal cord immobility and found that 31.3% exhibited penetration and 23.4% demonstrated aspiration on videofluoroscopy. Dysphagia in patients with vocal cord paralysis may be attributable to loss of pharyngeal innervation or to lower respiratory tract aspiration caused by glottic incompetence and diminished cough impulse [17].
Early recognition of the signs and symptoms of dysphagia can decrease the associated morbidity. The results of our study revealed poor oral intake and an altered cry/ voice after extubation to be the most sensitive predictors of dysphagia. Coughing during feeding and GER were found to be the most specific indicators for swallowing dysfunction in our cohort of patients. Our study findings match the results of similar studies conducted on dysphagia in other chronically ill pediatric populations [10, 18, 19]. However, before we did this study, poor oral intake was used as the primary marker for dysphagia during the postoperative period. A number of other symptoms contributing to poor oral intake were used to a lesser degree. Our results not only emphasize the need of bedside feeding assessments of patients who are at risk for dysphagia, but also point out the need to educate bedside caregivers so that they can easily recognize the signs and symptoms of dysphagia.
Pediatric patients undergoing open heart procedures are at risk for dysphagia. Therefore, dysphagia should be included in the list of surgical risks discussed with families when obtaining consent for operation. Perioperative management strategies beginning with the preoperative medical history and physical examination should aim to identify high-risk patients [20]. Intraoperative strategies should include the use of the new mini-multiplane TEE probes in neonates and smaller infants weighing less than 5.5 kg. Health care providers should consider using epicardial echocardiography in lieu of TEE for infants who weigh less than 3.5 kg.
Postoperative management should include early extubation when feasible and prompt recognition of the signs and symptoms of dysphagia with immediate referral to a speech therapist. On the basis of medical history, physical examination, and feeding presentation, recommendations for further workup can include a videofluoroscopic swallow study. This study is considered the gold standard for the evaluation of dysphagia. Tests for tolerance of various food consistencies have been developed for safe and effective swallowing [21]. If aspiration is seen, vocal cord paralysis should be suspected and the otolaryngology service, consulted. Strategies for postoperative management should incorporate a team approach to safely and efficiently maximize the patient's nutritional status. Recommended feeding regimens can include thickened liquids, nasogastric tube feedings, or the development of oral motor skills to assist with the swallowing process. Realistic short-term and long-term goals should be developed in conjunction with the primary caregiver with regularly scheduled follow-up appointments to evaluate compliance with the plan of care. By implementing these management strategies, the incidence, morbidity, and mortality associated with dysphagia can be decreased.
This study has several limitations. First, the small sample size decreased the power of the study and may have affected the outcomes in terms of the true incidence of dysphagia in this cohort and the predictors of swallowing dysfunction such as the use of TEE. Because there was no control group, the actual effect of TEE could not be assessed. The large number of study patients with complex lesions and the number of patients less than 3 years old may have contributed to a higher incidence of dysphagia in this study than in adult studies.
| Conclusions |
|---|
|
|
|---|
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Cote and A. Denault Transesophageal echocardiography-related complications: [Complications associees a l'echocardiographie transoesophagienne] Can J Anesth, September 1, 2008; 55(9): 622 - 647. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. A. Clement, H. El-Hakim, E. Z. Phillipos, and J. J. Cote Unilateral Vocal Cord Paralysis Following Patent Ductus Arteriosus Ligation in Extremely Low-Birth-Weight Infants Arch Otolaryngol Head Neck Surg, January 1, 2008; 134(1): 28 - 33. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Guillemaud, H. El-Hakim, S. Richards, and N. Chauhan Airway Pathologic Abnormalities in Symptomatic Children With Congenital Cardiac and Vascular Disease Arch Otolaryngol Head Neck Surg, July 1, 2007; 133(7): 672 - 676. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. Skinner, L. A. Halstead, C. S. Rubinstein, A. M. Atz, D. Andrews, and S. M. Bradley Laryngopharyngeal dysfunction after the Norwood procedure J. Thorac. Cardiovasc. Surg., November 1, 2005; 130(5): 1293 - 1301. [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 |