Abstract
The results of this study can inform the further investigation of the use of the side-lying position as a feeding treatment protocol for infants with vocal cord paresis (VCP).
Approximately 3% of term infants who undergo cardiovascular surgery experience unilateral vocal cord paresis (VCP) after surgery. Among preterm infants, this number is higher, ranging from 11% (Henry et al., 2019) to 43% among extremely low-birth-weight preterm infants (Benjamin et al., 2010). Other documented risks for VCP are prolonged intubation (Halan et al., 2017) and cervical extracorporeal membrane oxygenation (ECMO) cannulation (Itagaki et al., 2007).
VCP is well documented to lead to feeding difficulties for infants. In a study by Richter et al. (2016), 48% of children with VCP had feeding difficulties compared with only 19% of children with normal vocal cord movement. Children with VCP also had higher rates of readmission for feeding difficulties after initial discharge. Feeding difficulties lead to poor weight gain, oral aversion, and frequent respiratory infections as a result of aspiration. Infants with VCP experience these sequelae because of the higher risk of aspiration resulting from the vocal cords’ inability to fully protect the airway during swallowing. These feeding difficulties are connected with an increase in gastrostomy tube placement in infants with VCP (Benjamin et al., 2010; Richter et al., 2016).
The arduous process of transitioning to full oral feeding has been studied in programs that transition patients from gastrostomy tube placement after cardiac surgery. Studies have shown that approximately 50% of infants with VCP will be full oral feeders at age 1 yr (Richter et al., 2016) but that infants requiring tube feedings for longer than 6 mo before age 1 yr are at higher risk for long-term feeding disorders and food-avoidant disorders (Krom et al., 2017).
No research data exist on treatments to improve feeding safety among children with VCP. Research has shown that feeding preterm infants without comorbidity in an elevated side-lying position has a positive impact on vital sign stability (Park et al., 2014). Thickening agents are often used to aid in swallowing safety and to reduce the amount and frequency of aspiration. Again, however, there has been little research into safe ways to thicken breast milk and, therefore, no consensus among physicians, nutritionists, hospitals, or therapists. This lack of consensus leaves many infants who require thickening agents receiving formula for oral feedings.
It is common to position infants with VCP on their side, with the working vocal cord on the bottom side. For example, infants with left VCP would be positioned on their right side. The theoretical rationale is that if the infant has laryngeal penetration while swallowing, gravity will pull the milk toward the working vocal cord to protect the airway. However, no research on using this position in infants with VCP has been done.
This study aims to address the identified gap in evidence for feeding therapy for infants with unilateral VCP. This retrospective study was conducted in a free-standing pediatric hospital setting and investigated whether feeding position or other patient characteristics led to successful oral feeding among infants with VCP.
At the study hospital, the occupational therapist completes a feeding competency demonstrating understanding of side-lying feeding and VCP. The study hospital has protocols to evaluate the vocal cords and consult with occupational therapy postsurgery for procedures that are known to have a risk for VCP. Typically, the evaluating therapist uses a side-lying feeding position unless there are special circumstances or strong parent preference. Throughout the hospitalization and before discharge, the occupational therapy team trains the infant’s caregivers, who are expected to demonstrate their understanding of feeding recommendations. However, infants with VCP who do not fit into these protocols are not automatically referred to occupational therapy, and feedings in side lying may not be initiated. In these cases, the infant would be referred to occupational therapy only if the infant demonstrated feeding difficulties or required thickened liquids on the basis of a video fluoroscopic swallowing study.
Method
Study Design
The study was a retrospective descriptive study of the characteristics of, feeding interventions with, and feeding outcomes of infants with a diagnosis of VCP. It was approved by the Children’s Mercy institutional review board.
Participants
The 67 participants were infants age <1 yr who were diagnosed with unilateral VCP by an otolaryngologist via laryngoscope while they were inpatients at Children’s Mercy Hospital between January 1, 2017, and December 31, 2018. There were no exclusion criteria.
Data Collection
For infants meeting the inclusion criteria, the following data were extracted from the electronic medical record: infants’ gestational age, postmenstrual age at discharge, procedure or surgery performed before diagnosis of VCP (cervical ECMO cannulation, patent ductus arteriosus [PDA] ligation, aortic arch reconstruction, or prolonged intubation of ≥2 wk), type of milk given at discharge, video fluoroscopic swallowing study completed, infant feeding position, thickening agent used, consistency of thickening agent at discharge, whether a feeding tube was present at discharge, length of stay, and readmission for feeding difficulties ≤3 mo of discharge.
Data Analysis
Descriptive statistics, including means, standard deviations, and frequencies, were run to describe the population. Before any further analysis, we removed the 16 infants who were not orally feeding from analysis because thickener use and milk type would be irrelevant in this population. Data analysis was conducted using a Pearson χ2 test to identify relationships between the use of thickener at discharge and the infant’s feeding position, the procedure that caused the VCP, and the infant’s gestational age (defined as either preterm or full term). We also used χ2 tests to compare readmission (yes or no) with milk type at discharge (formula, breast milk, or both) and the infant’s gestational age. Finally, an analysis of variance test was performed to compare length of stay between groups on the basis of predisposing procedure.
Results
The final sample included 67 infants. Thirty-nine had aortic arch repair, 13 had PDA ligation, 12 had prolonged intubation, and 1 had ECMO cannulation before diagnosis of unilateral VCP. Two infants experienced VCP for unknown reasons. Twenty infants had a video fluoroscopic swallowing study completed.
For 16 infants, oral feeding was found to be unsafe in any position because of aspiration, leaving 51 infants for additional analysis. Forty-four infants were fed in a side-lying position, 6 were fed in an upright position, and 1 was fed in a semireclined position. Thirty-seven required a combination of tube feeding and oral feeding at discharge, 16 received all nutrition by tube, and 14 were fed solely orally. Of the infants feeding orally, 41 were born at term, and 31 did not require thickening agents at discharge. Of the 20 infants for whom thickening agents were used, 10 required honey-consistency liquids, 9 required nectar-consistency liquids, and 1 required spoon-thick-consistency liquids (Table 1). The mean gestational age of the infants was 37 wk (SD = 5.3), and the mean postmenstrual age at discharge was 50 wk (SD = 10.1). The mean length of stay was 53.5 days (SD = 47.9).
Participant Characteristics
Note. N = 67. ECMO = extracorporeal membrane oxygenation; NPO = nothing by mouth; PDA = patent ductus arteriosus.
Positioning and Thickener Use
Because only 1 infant was fed in a semireclined position, the data for this infant were removed before the χ2 analysis. Fewer infants fed in a side-lying position required thickening agents at discharge (30%), compared with 83% of infants who were fed in an upright position. We found a statistically significant difference between feeding position and thickener use, χ2(1, N = 50) = 6.630, p = .010.
Procedure and Thickener Use
Data for the 2 infants with an unknown procedure and the 1 infant with ECMO cannulation were removed before this analysis. Of the 51 infants who were feeding orally at discharge, 7 of 10 (70%) who underwent PDA ligation were using thickening agents, compared with 5 of 10 (50%) who had prolonged intubation and 7 of 31 infants (23%) who underwent arch repair. These results showed a statistically significantly difference, χ2(2, N = 49) = 8.039, p = .018.
Gestational Age and Thickener
Of the 51 infants eating orally at discharge, 10 were preterm, and 41 were term. We found that infants born preterm with VCP (70%) were more likely than infants born term with VCP (29%) to require thickening agents at discharge. This finding was statistically significant, χ2(1, N = 51) = 5.706, p = .017.
Procedure and Length of Stay
Infants who had aortic arch repair had a mean length of stay of 37.7 days (SD = 19.6); those who had prolonged intubation had a mean length of stay of 56.6 days (SD = 56.9); and those who had PDA ligation had a mean length of stay of 36.5 days (SD = 31.3). These differences were not statistically significant, F(2, 46) = 1.33, p = .275.
Milk Type and Readmission
We found no statistically significant differences in readmissions when comparing the types of milk infants were ingesting orally, χ2(2, N = 51) = 0.654, p = .721. Approximately 25% of infants who were orally ingesting formula were readmitted compared with 24% of infants who were orally ingesting breast milk.
Term Status and Readmission
More infants born preterm (60%) with VCP were readmitted for oral feeding difficulties than infants born term with VCP (15%). This difference was statistically significant, χ2(1, N = 51) = 9.195, p = .002.
Discussion
This study found that infants fed in a side-lying position were more likely than infants fed upright to be fed orally without a thickening agent. This finding suggests that infants fed in a side-lying position likely had lower frequencies and amount of aspiration. Of the 51 infants, 44 were fed in a side-lying position. The remaining 7 infants were fed in other positions because of barriers such as physician education and lack of referrals for feeding therapy. These barriers pose a potential confounding variable because of the delayed involvement of occupational therapy. There is no consensus on the safety of thickening breast milk, but these results suggest that feeding infants in the side-lying position may allow them to be breastfed safely or to be fed expressed breast milk if their caregiver prefers. This information is important for nurses, other medical providers, and therapists, because it provides guidance on the safest feeding position for infants with VCP.
The study also showed that patient characteristics may affect feeding success. Infants who had PDA ligation and prolonged intubation were more likely to require thickening agents and were also more likely to be readmitted for feeding difficulties. Similarly, infants who were born preterm were also more likely to require thickening agents and to be readmitted for feeding difficulties. Although potential confounding factors exist, such as that preterm infants were more likely to undergo a PDA ligation, the number of preterm infants in this study (n = 10) was too small to allow for more detailed analysis. In addition, PDA ligation was performed for both preterm and term infants in this study. Although additional research is still needed, this information allows medical providers and treating therapists to be more cautious with infants with these risk factors and to refer them appropriately for video fluoroscopic swallowing studies and close outpatient monitoring. Awareness of these risk factors also allows the medical team and treating therapists to better prepare families for what the journey ahead may look like for their infant.
It has been well documented infants with VCP experience an increased risk of feeding difficulties. Early feeding difficulties put infants at risk for long-term feeding difficulties and potentially for tube dependency. The goal of this single-center retrospective study was to investigate infant feeding treatments and patient factors affecting successful oral feeding and readmissions. This information is valuable to the medical and therapy teams to help families plan for the future for their infant experiencing unilateral VCP, especially if their infant is premature or has had a PDA ligation. This information could also improve the safety of oral feeding and decrease aspiration by using the side-lying position in feeding therapy and training caregivers to safely use this position at discharge. This is the only study to date to investigate feeding position in infants with VCP.
Study Limitations
The sample size was limited and from a single site, which may limit the generalizability of the findings. A larger multisite study would be beneficial in providing more robust evidence on feeding position and oral feeding best practices, as well as patient characteristics that may affect feeding success after unilateral VCP.
Implications for Occupational Therapy Practice
The primary purpose of this study was to investigate the impact of feeding position on feeding success. This preliminary study showed that feeding in the side-lying position with the working vocal cord down correlated with less thickener use. This study also sought to investigate other patient characteristics that led to feeding success and found that infants with history of PDA ligation and prematurity were more likely to need thickening agents and to be readmitted for feeding difficulties in our setting. This study has the following implications for occupational therapy practice: Practitioners should continue to feed infants with unilateral VCP in the side-lying position with the working vocal cord down. Practitioners should be aware that infants with history of PDA ligation or prematurity may be at higher risk of feeding difficulties and provide families with appropriate education before discharge. Practitioners should have a low threshold for completing a video fluoroscopic swallowing study on infants with unilateral VCP, especially infants with a history of PDA ligation or prematurity, because we found that they were more likely to require thickening agents, indicating they are more likely to be experiencing swallow dysfunction that could lead to their readmission for feeding difficulties. Completing a video fluoroscopic swallowing study before initial discharge could decrease risk for readmission.
Conclusion
The findings of this study demonstrate that feeding position affects feeding success in infants with unilateral VCP and, in particular, that feeding in a side-lying position with the unaffected vocal cord down may improve an infant’s ability to successfully orally feed without the need for thickening agents. This study’s findings support the need for further study in this area.
Footnotes
Acknowledgments
We thank Mary Verner in Data Intelligence at Children’s Mercy Hospital for her assistance with our initial data report.