- Email: [email protected]
Long-term follow-up of vocal fold movement impairment and feeding after neonatal cardiac surgery
International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Long-term follow-up of vocal fold movement impairment and feeding after neonatal cardiac surgery Amy Li Richter a,b,*, Julina Ongkasuwan a,b, Elena C. Ocampo c,d a
Bobby R. Alford Department of Otolaryngology—Head and Neck Surgery, Baylor College of Medicine, Houston, TX, United States Department of Pediatric Otolaryngology, Texas Children’s Hospital, Houston, TX, United States c Department of Pediatrics, Baylor College of Medicine, Houston, TX, United States d Department of Pediatric Cardiology, Texas Children’s Hospital, Houston, TX, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 2 January 2016 Received in revised form 11 February 2016 Accepted 13 February 2016 Available online 22 February 2016
Objective: To determine the long-term prognosis of children with vocal fold mobility impairment (VFMI) after cardiac surgery, with respect to time to normal feeding and incidence of admissions for pneumonia and feeding difficulties. Methods: A retrospective chart review was conducted of all neonates who had otolaryngology exam after cardiac surgery at a tertiary children’s hospital from May 2007 to May 2008. Charts were reviewed for demographics, type of cardiac surgery, vocal fold mobility, diet at time of discharge and at last followup, time to full oral feeding, and hospital admissions. Results: There were a total of 94 patients included in the study, 17 of whom had VFMI. While significantly more patients with VFMI required modified diet at discharge, 48% compared to 19% of patients with normal vocal fold mobility; there was no statistically significant difference in time to regular diet on long-term follow-up, 0.8 years (VFMI) compared to 0.4 years (normal vocal fold mobility). Of the 25 patients with modified diet or gastrostomy tube at discharge, 52% returned to full feeds within a year. There was no difference in hospitalizations for pneumonia in patients with or without VFMI. However in patients with VFMI, 35% required readmission for feeding difficulty or poor weight gain compared to only 5% in the infants with normal vocal fold mobility. Conclusion: After neonatal cardiac surgery, there do not appear to be long-term effects of VFMI with regards to readmission for pneumonia. However, there is an increased risk for hospitalization with respect to feeding difficulties in those neonates with VFMI. The overall prognosis for time to oral feeding is good. ß 2016 Published by Elsevier Ireland Ltd.
Keywords: Neonatal cardiac surgery Vocal fold movement impairment Feeding Dysphagia
1. Introduction Injury to the recurrent laryngeal nerve and vocal fold movement impairment (VFMI) is a known risk following cardiac surgery. The risk ranges from 1.7 to 67% depending on type of cardiac surgery, with increased risk in procedures requiring median sternotomy or manipulation of the aortic arch and in extremely low birth weight infants [1]. The recurrent laryngeal nerve runs a complicated course from the skull base, looping around the aortic arch on the left and the subclavian artery on the right before entering the larynx at the cricoarytenoid joint [2].
* Corresponding author at: Bobby R. Alford Department of Otolaryngology—Head and Neck Surgery, Baylor College of Medicine, Houston, TX, United States. Tel.: +1 832 822 3267; fax: +1 832 825 3251. E-mail address: [email protected] (A.L. Richter). http://dx.doi.org/10.1016/j.ijporl.2016.02.014 0165-5876/ß 2016 Published by Elsevier Ireland Ltd.
In adults with VFMI, over half of patients will report alterations in swallowing function [3]. In children, it appears that VFMI is associated with increased incidence of swallowing difficulties and prolonged hospitalization [4–6]. Additionally, injury to the recurrent laryngeal nerve and subsequent VFMI may place the neonate at increased risk for aspiration and pneumonia due to failure to adequately protect the airway. Some neonates may require thickened consistency of feeds or gastrostomy tube placement to decrease risk of aspiration [7,8]. Retrospective studies examining the long-term follow-up of VFMI suggest that 35–71% of neonates with VFMI after cardiac surgery will recover vocal fold movement depending on etiology of VFMI [8,9]. Dewan et al. reported VFMI outcomes following congenital cardiac surgery and correlated type of cardiac surgery, particularly manipulation of the aortic arch, with a high incidence of VFMI of 26.9%. This complication was associated with prolonged hospitalization and modified diet at time of discharge. There is
A.L. Richter et al. / International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214
212
Fig. 1. (A) Follow-up of patients with normal vocal fold mobility, n = 77. (B) Follow-up of patients with VFMI, n = 17.
limited data regarding the long-term prognosis of feeding difficulties after cardiac surgery. The purpose of this study is to report the long-term outcomes of the cohort from the Dewan et al. study to determine the impact of VFMI on long-term feeding outcomes and time to normal feeding after congenital heart surgery and morbidity of VFMI with respect to inpatient hospitalizations for aspiration, pneumonia, and feeding difficulties.
laryngoscopy was not available. Hospitalizations for pneumonias were indicated by clinical symptoms and findings on chest roentgenogram. Hospitalizations for feeding difficulties and poor weight gain were indicated by admission diagnoses and poor performance on the growth curve. Recurrent hospitalizations for further cardiac surgery were excluded. 2.2. Data analysis
2. Materials and methods 2.1. Data collection A retrospective chart review was conducted of all neonates who underwent flexible laryngology by otolaryngologist as part of a feeding evaluation protocol after cardiac surgery at Texas Children’s Hospital from May 2007 to May 2008. This is a follow-up study on the cohort of neonates reported by Dewan et al. [4]. Charts were reviewed until latest follow-up until December 2014. The institutional review board at Baylor College of Medicine approved this study. A total of 94 patients underwent cardiac surgery during this time period. All neonates were evaluated with flexible laryngoscopic evaluation following extubation. If the laryngoscopic findings were abnormal, patients underwent swallow evaluation and modified barium swallow study. Diet recommendations were made by a speech therapist. Inpatient charts from initial hospitalization for cardiac surgery were reviewed for demographic data, type of cardiac surgery performed (specifically noting involvement of aortic arch manipulation), vocal fold function, results of modified barium swallow (MBS) examination by a speech pathologist (if performed), and diet at time of discharge from initial cardiac procedure. Surgeries that involve aortic arch included: aortic arch advancement, Norwood procedure, Blalock–Taussig Shunt, Damus–Kaye–Stansel operation, patent ductus arteriosus ligation, coarctation of the aorta repair, and aortic arch reconstruction. The Norwood procedure recreates a neo-aorta and aortic arch using the main pulmonary artery in patients with hypoplastic left heart syndrome. The Blalock–Taussig Shunt creates a tube-like shunt between the subclavian artery to the pulmonary arteries to increase pulmonary circulation in patients with cyanotic heart defects such as tetralogy of Fallot. The Damus–Kaye–Stansel operation connects the main pulmonary arterial trunk to the aorta in neonates with obstructed systemic outflow in double inlet left ventricle. Diet at time of discharge included full oral feeds, modified thickened oral diet, partial oral diet with gastric feed supplementations, or exclusive gastric feeds (either nasogastric or gastrostomy tube). Outpatient charts were further reviewed for time to full oral feeds and incidences of inpatient hospitalizations for aspiration, pneumonia, and feeding difficulties. Outpatient follow-up visits were primarily documented by the congenital cardiac team and information on voice evaluation, swallow assessment, and flexible fiberoptic
Data analysis was performed with assistance from the Texas Children’s Hospital Outcomes and Impact Service. Continuous variables, including age, length of hospital stay, time to follow-up, and time to regular diet, were analyzed using one-tailed Student ttest and reported as medians with a minimum-maximum range or means with standard deviation. Categorical variables, including vocal fold mobility and readmissions, were reported with frequencies and percentages. Statistical analysis was performed using Student t-test, Fisher exact test, x2 test, logistic regression and multivariate analysis. 3. Results Ninety-four patients were included in the study, 17 of whom had vocal fold immobility. The mean time of follow up was 3.8 2.4 years for those with VFMI and 4.3 2.2 years for those with normal vocal fold mobility (p > 0.05). Nine patients died from causes unrelated to swallowing or feeding difficulty, of which one patient had VFMI. Eight patients were lost to follow-up due to reasons unrelated to vocal fold mobility (Fig. 1). Infants with genetic comorbidity were more likely to have VFMI (Table 1). Twenty-four neonates (31%) had modified diet at time of discharge by thickened feeds, nasogastric, or gastrostomy feeding, of which fifteen patients Table 1 Patient characteristics at time of discharge from initial surgery. Characteristic
VFMI
Normal VF mobility
p
n Age at time of surgery (days) Genetic comorbidity Length of hospital stay (days) Aortic arch manipulation* Aspiration on modified barium swallow (MBS)** Abnormal cough or gag reflex Regular diet at discharge
17 (18%) 8.0 6.0 7 (41%) 34.9 29.5 93% 47%
77 (82%) 10.4 6.7 16 (21%) 22.9 14.6 63% 16%
NS <0.01 0.02 0.02
73% 52%
32% 81%
0.01 <0.01
NS Not significant. * Arch manipulation surgeries include aortic arch advancement, Norwood procedure, Blalock–Taussig Shunt, Damus–Kaye–Stansel operation, patent ductus arteriosus ligation, coarctation of the aorta repair, and aortic arch reconstruction. ** Not all patients received evaluation with MBS, only those with abnormal laryngoscopy findings.
A.L. Richter et al. / International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214 Table 2 Patient characteristics at time of most recent follow-up. Characteristic
VFMI
Normal VF mobility
p
Mean time to most recent follow-up (years) Mean time to regular diet, all patients (years) Mean time to regular diet, modified diet at discharge (years) Readmissions for aspiration-related pneumonia Readmission feeding difficulty and poor weight gain Genetic comorbidity and modified diet at most recent follow-up Lost to follow up Deceased
3.8 2.4
4.3 2.2
NS
0.8 1.4
0.4 1.1
NS
1.3 1.7
1.8 1.1
NS
1 (6%)
9 (12%)
NS
6 (35%)
4 (5%)
<0.01
0
3 (4%)
NS
1 (6%) 1 (6%)
8 (10%) 8 (10%)
NS NS
NS Not significant.
(63%) had normal vocal fold function (Table 1). Of the 17 patients with VFMI, 9 neonates had modified diet at time of discharge and the mean time to regular diet was 1.3 1.67 years. Of the 77 patients with normal vocal fold mobility, 15 had modified diet at time of discharge and the mean time from discharge to regular diet was 1.8 1.9 years (Table 2). The diet at time of discharge was impacted by vocal fold mobility (p = 0.02) but the mean time from discharge to regular diet was not affected by VFMI (p > 0.05). In the 25 patients with modified diet or gastrostomy tube at discharge, 52% returned to full oral feeds within a year. Two patients still had a gastrostomy tubes at most recent follow-up. Genetic comorbidity did not affect time to regular diet or necessity for prolonged gastrostomy tube (Table 2). In the entire cohort, there were 10 (13%) hospitalizations for aspiration and pneumonia, of which only one patient had VFMI (p > 0.05) (Table 2). There was an increased risk of readmission for feeding difficulty and poor weight gain in infants with VFMI (35%) compared to patients with normal vocal fold mobility (5%) (p < 0.01) (Table 2) The most common reasons for readmission were viral upper respiratory infection and gastrointestinal illness, excluding readmissions for revision cardiac procedures. 4. Discussion Injury to the recurrent laryngeal nerve during congenital cardiac surgery is common and VFMI impacts neonatal recovery, contributing to increased length of hospitalization and feeding restriction at time of discharge [4]. Dewan et al. reported an overall incidence of VFMI for all types of cardiac surgery at time of discharge of 18% [4]. There was a higher incidence of VFMI (26.9%) in patients undergoing cardiac surgery with manipulation of the aortic arch. The type of cardiac surgery was the most significant factor for prediction of VFMI. These neonates were more likely to suffer from abnormal cough– gag reflex, modified diet at discharge and need for tube feeding, and increased length of hospitalization. Our study provides the long-term follow-up for this cohort. Sachdeva et al. further elucidated the relationship between VFMI and feeding difficulties. In a study of 38 neonates with VFMI after cardiac surgery, every neonate with VFMI who underwent a swallow evaluation was found to have evidence of aspiration, penetration, or discordant swallowing. Of this group, only 13% were tolerating full oral feeds at time of discharge and most patients required diet modification or enteral feeds. In a limited follow up time of six months in some patients in this cohort, 70% of patients demonstrated normal or improved swallowing [7]. Several studies have found increased incidence (23–40%) of VFMI in low and extremely low birth weight infants after PDA
213
ligation [1,10]. Infants with vocal fold dysfunction were more likely to require prolonged mechanical ventilation, develop chronic lung disease, require enteral feeds, and undergo Nissen fundoplication when compared to infants without vocal fold dysfunction [1]. There is limited data regarding long-term follow up of neonates with VFMI after cardiac surgery regarding vocal fold mobility recovery and feeding outcomes. Truong et al. found that in a group of 80 neonates with VFMI, 35% of the patients had spontaneous recovery of vocal fold mobility at a mean rate of 6.6 months. The authors suggest that there may be a higher rate of recovery, as patients with asymptomatic VFMI after cardiac surgery may not be detected in this retrospective review. They also noted that children with VFMI had a high rate of subsequent surgical intervention of 27%, most frequently gastrostomy tube placement (15%), followed by tracheostomy (4%), and vocal fold medialization procedure (8%) [8]. We found the time to regular oral feeding was less than two years in most infants who were on modified feeding at time of discharge, independent of vocal fold mobility (Table 2). The incidence of admission for aspiration-related pneumonia did not appear to be impacted by VFMI (Table 2). This may be due to increased awareness of aspiration risk in patients with VFMI and improved discharge counseling with strict diet recommendations to avoid aspiration and pneumonia. However, VFMI does appear to impact long-term risk for inpatient hospitalization feeding difficulties as indicated by poor weight gain and poor performance on the growth curve. The most common reasons for hospital admissions include viral upper respiratory infection and gastrointestinal illness. Infants with genetic comorbidities appear to have increased risk for VFMI after cardiac surgery and modified diet at time of initial discharge, but the presence of genetic comorbidity does not appear to affect time to regular diet. This information can be useful when counseling families regarding expectations for risk of aspiration and pneumonia and expectations for feeding and progression along the growth curve after discharge and what can be expected regarding prognosis for regular diet. Congenital vocal fold paresis cannot be ruled out in this cohort, as patients did not undergo preoperative laryngoscopic evaluation. This study is also limited by its retrospective study design. Not all patients received a follow-up laryngoscopic exam or swallow evaluation so the incidence and rate of recovery of VFMI is unknown in this group. Future studies are needed to follow these VFMI infants with serial examinations by otolaryngology in order to assess the true incidence and rate of recovery of VFMI and the direct correlation with return to regular diet. 5. Conclusion Neonates with congenital heart defects often have a difficult postoperative course and require frequent readmissions for a variety of reasons. In this long-term follow up study of neonates with VFMI after cardiac surgery, there does not appear to be increased risk of admission for aspiration and pneumonia. On the other hand, VFMI does appear to increase risk for readmission for feeding difficulties or poor weight gain. The overall prognosis for time to oral feeding is good with most infants being orally feed within two years. Conflict of interest statement The authors, Amy Richter, Julina Ongkasuwan, and Elena Ocampo, have no financial interests or conflicts of interest to disclose.
214
A.L. Richter et al. / International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214
References [1] J.R. Benjamin, P.B. Smith, C.M. Cotton, J. Jaggers, R.F. Goldstein, W.F. Malcolm, Long-term morbidities associated with vocal fold paralysis after surgical closure of a patent ductus arteriosus in extermely low birth weight infants, J. Perinatol. 30 (2010) 408–413. [2] P.W. Flint, B.H. Haughey, V.J. Lung, J.K. Niparko, M.A. Richardson, K.T. Robbins, et al., Cummings Otolaryngology Head and Neck Surgery, fifth ed., Mosby Elsevier, Medical Pub. Division, Philadelphia, 2010. [3] B. Ollivere, K. Duce, G. Rowlands, P. Harrison, B.J. O’Reilly, Swallowing dysfunction in patients with unilateral vocal fold paralysis: aetiology and outcomes, J. Laryngol. Otol. 120 (2006) 38–41. [4] K. Dewan, C. Cephus, V. Owcharzak, E. Ocampo, Incidence and implication of vocal fold paresis following neonatal cardiac surgery, Laryngoscope 122 (2012) 2781–2785. [5] L. Kohr, M. Dargan, A. Hague, S. Nelson, E. Duffy, C. Backer, et al., The incidence of dysphagia in pediatric patients after open heart procedures with transesophageal echocardiography, Ann. Thorac. Surg. 76 (2003) 1450–1456.
[6] M. Skinner, L. Halstead, C. Rubinstein, D. Atz Am Andrews, S. Bradley, Laryngophargeal dysfunction after the Norwood procedure, J. Thorac. Cardiovasc. Surg. 130 (2005) 1293–1301. [7] R. Sachdeva, E. Hussain, M.M. Moss, M.L. Schmitz, R.M. Ray, M. Imamura, et al., Vocal cord dysfunction and feeding difficulties after pediatric cardiovascular surgery, J. Pediatr. 151 (2007) 312–315. [8] M.T. Truong, A.H. Messner, J.E. Kerschner, M. Scholes, J. Wong-Dominguez, H.A. Milczuk, et al., Pediatric vocal fold paralysis after cardiac surgery: rate of recovery and sequelae, Otolaryngol. Head Neck Surg. 137 (2007) 780–784. [9] H. Daya, A. Hosni, I. Bejar-Solar, J. Evans, M. Bailey, Pediatric vocal fold paralysis: a long-term retrospective study, Arch. Otolaryngol. Head Neck Surg. 126 (2000) 21–25. [10] R. Zbar, A. Chen, D. Behrendt, E. Bell, R. Smith, Incidence of vocal fold paralysis in infants undergoing ligation of patent ductus arteriosus, Ann. Thorac. Surg. 61 (1996) 814–816.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Long-term follow-up of vocal fold movement impairment and feeding after neonatal cardiac surgery Amy Li Richter a,b,*, Julina Ongkasuwan a,b, Elena C. Ocampo c,d a
Bobby R. Alford Department of Otolaryngology—Head and Neck Surgery, Baylor College of Medicine, Houston, TX, United States Department of Pediatric Otolaryngology, Texas Children’s Hospital, Houston, TX, United States c Department of Pediatrics, Baylor College of Medicine, Houston, TX, United States d Department of Pediatric Cardiology, Texas Children’s Hospital, Houston, TX, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 2 January 2016 Received in revised form 11 February 2016 Accepted 13 February 2016 Available online 22 February 2016
Objective: To determine the long-term prognosis of children with vocal fold mobility impairment (VFMI) after cardiac surgery, with respect to time to normal feeding and incidence of admissions for pneumonia and feeding difficulties. Methods: A retrospective chart review was conducted of all neonates who had otolaryngology exam after cardiac surgery at a tertiary children’s hospital from May 2007 to May 2008. Charts were reviewed for demographics, type of cardiac surgery, vocal fold mobility, diet at time of discharge and at last followup, time to full oral feeding, and hospital admissions. Results: There were a total of 94 patients included in the study, 17 of whom had VFMI. While significantly more patients with VFMI required modified diet at discharge, 48% compared to 19% of patients with normal vocal fold mobility; there was no statistically significant difference in time to regular diet on long-term follow-up, 0.8 years (VFMI) compared to 0.4 years (normal vocal fold mobility). Of the 25 patients with modified diet or gastrostomy tube at discharge, 52% returned to full feeds within a year. There was no difference in hospitalizations for pneumonia in patients with or without VFMI. However in patients with VFMI, 35% required readmission for feeding difficulty or poor weight gain compared to only 5% in the infants with normal vocal fold mobility. Conclusion: After neonatal cardiac surgery, there do not appear to be long-term effects of VFMI with regards to readmission for pneumonia. However, there is an increased risk for hospitalization with respect to feeding difficulties in those neonates with VFMI. The overall prognosis for time to oral feeding is good. ß 2016 Published by Elsevier Ireland Ltd.
Keywords: Neonatal cardiac surgery Vocal fold movement impairment Feeding Dysphagia
1. Introduction Injury to the recurrent laryngeal nerve and vocal fold movement impairment (VFMI) is a known risk following cardiac surgery. The risk ranges from 1.7 to 67% depending on type of cardiac surgery, with increased risk in procedures requiring median sternotomy or manipulation of the aortic arch and in extremely low birth weight infants [1]. The recurrent laryngeal nerve runs a complicated course from the skull base, looping around the aortic arch on the left and the subclavian artery on the right before entering the larynx at the cricoarytenoid joint [2].
* Corresponding author at: Bobby R. Alford Department of Otolaryngology—Head and Neck Surgery, Baylor College of Medicine, Houston, TX, United States. Tel.: +1 832 822 3267; fax: +1 832 825 3251. E-mail address: [email protected] (A.L. Richter). http://dx.doi.org/10.1016/j.ijporl.2016.02.014 0165-5876/ß 2016 Published by Elsevier Ireland Ltd.
In adults with VFMI, over half of patients will report alterations in swallowing function [3]. In children, it appears that VFMI is associated with increased incidence of swallowing difficulties and prolonged hospitalization [4–6]. Additionally, injury to the recurrent laryngeal nerve and subsequent VFMI may place the neonate at increased risk for aspiration and pneumonia due to failure to adequately protect the airway. Some neonates may require thickened consistency of feeds or gastrostomy tube placement to decrease risk of aspiration [7,8]. Retrospective studies examining the long-term follow-up of VFMI suggest that 35–71% of neonates with VFMI after cardiac surgery will recover vocal fold movement depending on etiology of VFMI [8,9]. Dewan et al. reported VFMI outcomes following congenital cardiac surgery and correlated type of cardiac surgery, particularly manipulation of the aortic arch, with a high incidence of VFMI of 26.9%. This complication was associated with prolonged hospitalization and modified diet at time of discharge. There is
A.L. Richter et al. / International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214
212
Fig. 1. (A) Follow-up of patients with normal vocal fold mobility, n = 77. (B) Follow-up of patients with VFMI, n = 17.
limited data regarding the long-term prognosis of feeding difficulties after cardiac surgery. The purpose of this study is to report the long-term outcomes of the cohort from the Dewan et al. study to determine the impact of VFMI on long-term feeding outcomes and time to normal feeding after congenital heart surgery and morbidity of VFMI with respect to inpatient hospitalizations for aspiration, pneumonia, and feeding difficulties.
laryngoscopy was not available. Hospitalizations for pneumonias were indicated by clinical symptoms and findings on chest roentgenogram. Hospitalizations for feeding difficulties and poor weight gain were indicated by admission diagnoses and poor performance on the growth curve. Recurrent hospitalizations for further cardiac surgery were excluded. 2.2. Data analysis
2. Materials and methods 2.1. Data collection A retrospective chart review was conducted of all neonates who underwent flexible laryngology by otolaryngologist as part of a feeding evaluation protocol after cardiac surgery at Texas Children’s Hospital from May 2007 to May 2008. This is a follow-up study on the cohort of neonates reported by Dewan et al. [4]. Charts were reviewed until latest follow-up until December 2014. The institutional review board at Baylor College of Medicine approved this study. A total of 94 patients underwent cardiac surgery during this time period. All neonates were evaluated with flexible laryngoscopic evaluation following extubation. If the laryngoscopic findings were abnormal, patients underwent swallow evaluation and modified barium swallow study. Diet recommendations were made by a speech therapist. Inpatient charts from initial hospitalization for cardiac surgery were reviewed for demographic data, type of cardiac surgery performed (specifically noting involvement of aortic arch manipulation), vocal fold function, results of modified barium swallow (MBS) examination by a speech pathologist (if performed), and diet at time of discharge from initial cardiac procedure. Surgeries that involve aortic arch included: aortic arch advancement, Norwood procedure, Blalock–Taussig Shunt, Damus–Kaye–Stansel operation, patent ductus arteriosus ligation, coarctation of the aorta repair, and aortic arch reconstruction. The Norwood procedure recreates a neo-aorta and aortic arch using the main pulmonary artery in patients with hypoplastic left heart syndrome. The Blalock–Taussig Shunt creates a tube-like shunt between the subclavian artery to the pulmonary arteries to increase pulmonary circulation in patients with cyanotic heart defects such as tetralogy of Fallot. The Damus–Kaye–Stansel operation connects the main pulmonary arterial trunk to the aorta in neonates with obstructed systemic outflow in double inlet left ventricle. Diet at time of discharge included full oral feeds, modified thickened oral diet, partial oral diet with gastric feed supplementations, or exclusive gastric feeds (either nasogastric or gastrostomy tube). Outpatient charts were further reviewed for time to full oral feeds and incidences of inpatient hospitalizations for aspiration, pneumonia, and feeding difficulties. Outpatient follow-up visits were primarily documented by the congenital cardiac team and information on voice evaluation, swallow assessment, and flexible fiberoptic
Data analysis was performed with assistance from the Texas Children’s Hospital Outcomes and Impact Service. Continuous variables, including age, length of hospital stay, time to follow-up, and time to regular diet, were analyzed using one-tailed Student ttest and reported as medians with a minimum-maximum range or means with standard deviation. Categorical variables, including vocal fold mobility and readmissions, were reported with frequencies and percentages. Statistical analysis was performed using Student t-test, Fisher exact test, x2 test, logistic regression and multivariate analysis. 3. Results Ninety-four patients were included in the study, 17 of whom had vocal fold immobility. The mean time of follow up was 3.8 2.4 years for those with VFMI and 4.3 2.2 years for those with normal vocal fold mobility (p > 0.05). Nine patients died from causes unrelated to swallowing or feeding difficulty, of which one patient had VFMI. Eight patients were lost to follow-up due to reasons unrelated to vocal fold mobility (Fig. 1). Infants with genetic comorbidity were more likely to have VFMI (Table 1). Twenty-four neonates (31%) had modified diet at time of discharge by thickened feeds, nasogastric, or gastrostomy feeding, of which fifteen patients Table 1 Patient characteristics at time of discharge from initial surgery. Characteristic
VFMI
Normal VF mobility
p
n Age at time of surgery (days) Genetic comorbidity Length of hospital stay (days) Aortic arch manipulation* Aspiration on modified barium swallow (MBS)** Abnormal cough or gag reflex Regular diet at discharge
17 (18%) 8.0 6.0 7 (41%) 34.9 29.5 93% 47%
77 (82%) 10.4 6.7 16 (21%) 22.9 14.6 63% 16%
NS <0.01 0.02 0.02
73% 52%
32% 81%
0.01 <0.01
NS Not significant. * Arch manipulation surgeries include aortic arch advancement, Norwood procedure, Blalock–Taussig Shunt, Damus–Kaye–Stansel operation, patent ductus arteriosus ligation, coarctation of the aorta repair, and aortic arch reconstruction. ** Not all patients received evaluation with MBS, only those with abnormal laryngoscopy findings.
A.L. Richter et al. / International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214 Table 2 Patient characteristics at time of most recent follow-up. Characteristic
VFMI
Normal VF mobility
p
Mean time to most recent follow-up (years) Mean time to regular diet, all patients (years) Mean time to regular diet, modified diet at discharge (years) Readmissions for aspiration-related pneumonia Readmission feeding difficulty and poor weight gain Genetic comorbidity and modified diet at most recent follow-up Lost to follow up Deceased
3.8 2.4
4.3 2.2
NS
0.8 1.4
0.4 1.1
NS
1.3 1.7
1.8 1.1
NS
1 (6%)
9 (12%)
NS
6 (35%)
4 (5%)
<0.01
0
3 (4%)
NS
1 (6%) 1 (6%)
8 (10%) 8 (10%)
NS NS
NS Not significant.
(63%) had normal vocal fold function (Table 1). Of the 17 patients with VFMI, 9 neonates had modified diet at time of discharge and the mean time to regular diet was 1.3 1.67 years. Of the 77 patients with normal vocal fold mobility, 15 had modified diet at time of discharge and the mean time from discharge to regular diet was 1.8 1.9 years (Table 2). The diet at time of discharge was impacted by vocal fold mobility (p = 0.02) but the mean time from discharge to regular diet was not affected by VFMI (p > 0.05). In the 25 patients with modified diet or gastrostomy tube at discharge, 52% returned to full oral feeds within a year. Two patients still had a gastrostomy tubes at most recent follow-up. Genetic comorbidity did not affect time to regular diet or necessity for prolonged gastrostomy tube (Table 2). In the entire cohort, there were 10 (13%) hospitalizations for aspiration and pneumonia, of which only one patient had VFMI (p > 0.05) (Table 2). There was an increased risk of readmission for feeding difficulty and poor weight gain in infants with VFMI (35%) compared to patients with normal vocal fold mobility (5%) (p < 0.01) (Table 2) The most common reasons for readmission were viral upper respiratory infection and gastrointestinal illness, excluding readmissions for revision cardiac procedures. 4. Discussion Injury to the recurrent laryngeal nerve during congenital cardiac surgery is common and VFMI impacts neonatal recovery, contributing to increased length of hospitalization and feeding restriction at time of discharge [4]. Dewan et al. reported an overall incidence of VFMI for all types of cardiac surgery at time of discharge of 18% [4]. There was a higher incidence of VFMI (26.9%) in patients undergoing cardiac surgery with manipulation of the aortic arch. The type of cardiac surgery was the most significant factor for prediction of VFMI. These neonates were more likely to suffer from abnormal cough– gag reflex, modified diet at discharge and need for tube feeding, and increased length of hospitalization. Our study provides the long-term follow-up for this cohort. Sachdeva et al. further elucidated the relationship between VFMI and feeding difficulties. In a study of 38 neonates with VFMI after cardiac surgery, every neonate with VFMI who underwent a swallow evaluation was found to have evidence of aspiration, penetration, or discordant swallowing. Of this group, only 13% were tolerating full oral feeds at time of discharge and most patients required diet modification or enteral feeds. In a limited follow up time of six months in some patients in this cohort, 70% of patients demonstrated normal or improved swallowing [7]. Several studies have found increased incidence (23–40%) of VFMI in low and extremely low birth weight infants after PDA
213
ligation [1,10]. Infants with vocal fold dysfunction were more likely to require prolonged mechanical ventilation, develop chronic lung disease, require enteral feeds, and undergo Nissen fundoplication when compared to infants without vocal fold dysfunction [1]. There is limited data regarding long-term follow up of neonates with VFMI after cardiac surgery regarding vocal fold mobility recovery and feeding outcomes. Truong et al. found that in a group of 80 neonates with VFMI, 35% of the patients had spontaneous recovery of vocal fold mobility at a mean rate of 6.6 months. The authors suggest that there may be a higher rate of recovery, as patients with asymptomatic VFMI after cardiac surgery may not be detected in this retrospective review. They also noted that children with VFMI had a high rate of subsequent surgical intervention of 27%, most frequently gastrostomy tube placement (15%), followed by tracheostomy (4%), and vocal fold medialization procedure (8%) [8]. We found the time to regular oral feeding was less than two years in most infants who were on modified feeding at time of discharge, independent of vocal fold mobility (Table 2). The incidence of admission for aspiration-related pneumonia did not appear to be impacted by VFMI (Table 2). This may be due to increased awareness of aspiration risk in patients with VFMI and improved discharge counseling with strict diet recommendations to avoid aspiration and pneumonia. However, VFMI does appear to impact long-term risk for inpatient hospitalization feeding difficulties as indicated by poor weight gain and poor performance on the growth curve. The most common reasons for hospital admissions include viral upper respiratory infection and gastrointestinal illness. Infants with genetic comorbidities appear to have increased risk for VFMI after cardiac surgery and modified diet at time of initial discharge, but the presence of genetic comorbidity does not appear to affect time to regular diet. This information can be useful when counseling families regarding expectations for risk of aspiration and pneumonia and expectations for feeding and progression along the growth curve after discharge and what can be expected regarding prognosis for regular diet. Congenital vocal fold paresis cannot be ruled out in this cohort, as patients did not undergo preoperative laryngoscopic evaluation. This study is also limited by its retrospective study design. Not all patients received a follow-up laryngoscopic exam or swallow evaluation so the incidence and rate of recovery of VFMI is unknown in this group. Future studies are needed to follow these VFMI infants with serial examinations by otolaryngology in order to assess the true incidence and rate of recovery of VFMI and the direct correlation with return to regular diet. 5. Conclusion Neonates with congenital heart defects often have a difficult postoperative course and require frequent readmissions for a variety of reasons. In this long-term follow up study of neonates with VFMI after cardiac surgery, there does not appear to be increased risk of admission for aspiration and pneumonia. On the other hand, VFMI does appear to increase risk for readmission for feeding difficulties or poor weight gain. The overall prognosis for time to oral feeding is good with most infants being orally feed within two years. Conflict of interest statement The authors, Amy Richter, Julina Ongkasuwan, and Elena Ocampo, have no financial interests or conflicts of interest to disclose.
214
A.L. Richter et al. / International Journal of Pediatric Otorhinolaryngology 83 (2016) 211–214
References [1] J.R. Benjamin, P.B. Smith, C.M. Cotton, J. Jaggers, R.F. Goldstein, W.F. Malcolm, Long-term morbidities associated with vocal fold paralysis after surgical closure of a patent ductus arteriosus in extermely low birth weight infants, J. Perinatol. 30 (2010) 408–413. [2] P.W. Flint, B.H. Haughey, V.J. Lung, J.K. Niparko, M.A. Richardson, K.T. Robbins, et al., Cummings Otolaryngology Head and Neck Surgery, fifth ed., Mosby Elsevier, Medical Pub. Division, Philadelphia, 2010. [3] B. Ollivere, K. Duce, G. Rowlands, P. Harrison, B.J. O’Reilly, Swallowing dysfunction in patients with unilateral vocal fold paralysis: aetiology and outcomes, J. Laryngol. Otol. 120 (2006) 38–41. [4] K. Dewan, C. Cephus, V. Owcharzak, E. Ocampo, Incidence and implication of vocal fold paresis following neonatal cardiac surgery, Laryngoscope 122 (2012) 2781–2785. [5] L. Kohr, M. Dargan, A. Hague, S. Nelson, E. Duffy, C. Backer, et al., The incidence of dysphagia in pediatric patients after open heart procedures with transesophageal echocardiography, Ann. Thorac. Surg. 76 (2003) 1450–1456.
[6] M. Skinner, L. Halstead, C. Rubinstein, D. Atz Am Andrews, S. Bradley, Laryngophargeal dysfunction after the Norwood procedure, J. Thorac. Cardiovasc. Surg. 130 (2005) 1293–1301. [7] R. Sachdeva, E. Hussain, M.M. Moss, M.L. Schmitz, R.M. Ray, M. Imamura, et al., Vocal cord dysfunction and feeding difficulties after pediatric cardiovascular surgery, J. Pediatr. 151 (2007) 312–315. [8] M.T. Truong, A.H. Messner, J.E. Kerschner, M. Scholes, J. Wong-Dominguez, H.A. Milczuk, et al., Pediatric vocal fold paralysis after cardiac surgery: rate of recovery and sequelae, Otolaryngol. Head Neck Surg. 137 (2007) 780–784. [9] H. Daya, A. Hosni, I. Bejar-Solar, J. Evans, M. Bailey, Pediatric vocal fold paralysis: a long-term retrospective study, Arch. Otolaryngol. Head Neck Surg. 126 (2000) 21–25. [10] R. Zbar, A. Chen, D. Behrendt, E. Bell, R. Smith, Incidence of vocal fold paralysis in infants undergoing ligation of patent ductus arteriosus, Ann. Thorac. Surg. 61 (1996) 814–816.