Abstract
Objective:
The gold standard for diagnosis of craniosynostosis is a clinical examination and motionless head computed tomography (CT). Computed tomography sedation is associated with increased cost, resource utilization, medical, and possible developmental risks. This study investigates whether a “feed and swaddle” protocol can be used to achieve diagnostic quality craniofacial imaging without the use of infant sedation.
Design:
Prospective cohort study.
Setting:
Tertiary academic medical center.
Patients:
Ninety patients <18 months of age undergoing evaluation for craniosynostosis from 2012 to 2018.
Interventions:
A feed and swaddle protocol.
Main Outcome Measures:
Diagnostic level imaging without the use of infant sedation.
Results:
Eighty-five (94%) achieved a diagnostic quality craniofacial CT scan using the “feed and swaddle” method. Mean patient age was 24.0 ± 10.0 weeks. Craniosynostosis was diagnosed in 74% of patients. Mean age of patients with successful completion of a CT scan was 23.7 ± 9.6 weeks, compared to 27.2 ± 17.1 weeks for unsuccessful completion. Mean weight for the successful group was 15.6 ± 2.9 pounds and 15.9 ± 2.5 pounds for the unsuccessful group. Mean travel distance was 59.2 ± 66.5 miles for successful patients and 66.5 ± 61.5 miles for unsuccessful patients. For the unsuccessful patients, there were no delays in surgical planning or scheduling.
Conclusion:
The “feed and swaddle” protocol described here is an effective alternative to infant sedation for motionless craniofacial CT imaging.
Introduction
Computed tomographic (CT) imaging in the pediatric population often requires sedation to obtain diagnostic images without motion artifact. In the past several years, concerns have been raised regarding a possible effect of sedation and anesthesia on neurodevelopment, especially in patients <3 years (Flick et al., 2011; Ing et al., 2014; Chinn et al., 2016; Andropoulos and Greene, 2017). Anesthesia in children, especially younger than 6 months, carries additional risk (including oral, intravenous, and inhalational medications; Cote and Wilson, 2006; Wachtel et al., 2009). Sedated imaging requires additional administrative coordination and specialty-trained personnel, is associated with increased cost and resource needs, and places additional emotional and time burden on patient families.
Alternatives to the use of sedation to achieve motionless imaging have been described using auditory or visual distraction, preprocedural education to diminish anxiety, soothing environmental surroundings, and modification of feeding to facilitate sleep (Sury et al., 2005; Koch, 2008; Raschle et al., 2009; Edwards and Arthurs, 2011; Windram et al., 2011; Koller and Goldman, 2012). The “feed and swaddle” technique (Mathur et al., 2007; Hansen, 2009; Fogel et al., 2011) has so far shown the greatest success in infants <3 months (Hansen, 2009).
The craniofacial disorders programming at our institution uses low-dose radiation, noncontrast craniofacial CT imaging to evaluate infants and young children for possible craniosynostosis (Didier et al., 2010; Zarella et al., 2016). Over 95% of children <1 year require sedation to perform the 30-second CT sequence. We investigated the use of the “feed and swaddle” technique as an alternative to sedation for elective craniofacial CT imaging in children <18 months at our institution. This study describes the implementation and outcomes of use of the feed and swaddle protocol.
Methods
This study was approved by the institutional review board. We conducted a prospective, single-center cohort study of all patients referred for craniofacial CT scans to evaluate for craniosynostosis between 2012 and 2018. CT imaging was indicated after initial consultation by the craniofacial team plastic surgeons, neurosurgeon, or pediatric nurse practitioners, and was used to confirm clinical findings. Families were then offered enrollment in the study and were also eligible to participate in the protocol without enrolling, although none chose to do so. Inclusion criteria was children 0 to 2 years old who had clinical examination findings of craniosynostosis and had not yet undergone surgery for craniosynostosis. Exclusion criteria included social or medical hardships that limited their ability participate. Families were provided with a brochure describing the “feed and swaddle” method (Hansen, 2009) and subsequently contacted via telephone by a Child Life Specialist to fully review the protocol and answer questions or concerns.
Feed and Swaddle Protocol
Children were kept NPO (nothing by mouth) for 4 hours prior to the scheduled time of the scan. A child life specialist met parents at the hospital and brought them to a quiet room (with chairs and blankets) adjacent to the CT scanner suite. The child was then fed their usual diet, swaddled, and allowed to fall asleep in the family/caregiver’s arms. Once asleep, the child was placed in the scanner and the CT images for the craniofacial protocol were obtained. All diagnostic quality scans were interpreted by an attending radiologist. A successful outcome was defined as the completion of a diagnostic quality CT scan without sedation using the “feed and swaddle” method. If the child was unable to lie still the scan was discontinued and subsequently scheduled with sedation.
Data Analysis
For the analysis, we generated descriptive statistics using the mean and standard deviation for continuous variables along with frequency and percentage for binary variables using Excel (Microsoft Corp, Richmond, Washington).
Results
A total of 90 patients were enrolled from 2012 to 2018 (Table 1). Mean patient age was 24.0 weeks (±10.0 weeks). Eighty-five (94%) patients achieved a diagnostic quality craniofacial CT scan using the “feed and swaddle” method (Table 2). The mean age of patients successfully completing the scan was 23.7 weeks (±9.6), compared to 27.2 weeks (±17.1) for unsuccessful patients. The oldest patient with a successful study was 60 weeks. Patient age was fairly evenly distributed between 7 and 60 weeks. The majority of patients were male (69%), white (87%), non-Hispanic (74%), and were born at term without significant comorbidities. Craniosynostosis was present in the CT scan in 74% of patients. The remaining (36%) of patients were found to have either plagiocephaly or a normal scan. Fifty-four percent of patients had one or more sibling. Mean weights for the 2 groups was equivalent at 16 pounds. Synostosis of one or more sutures was diagnosed in 65 (76%) of the successful patients and 2 (40%) of the unsuccessful patients. Mean travel distance was similar for the successful patients at 59.2 miles (± 66.5), compared to 66.5 miles (± 61.5) for the unsuccessful group. Due to the disproportionate number of patients who were successful, we were unable to generate meaningful statistical comparisons between the 2 groups. Of the 5 unsuccessful patients, 2 were fed and swaddled in a noisy environment, and 1 patient was only kept NPO for 2 rather than 4 hours, and thus appeared to be insufficiently tired after eating. The other 2 patients followed the protocol and were still not able to fall asleep.
Patient Characteristics (n = 90).a
Abbreviation: SD, standard deviation.
aDue to rounding, percentages may not always appear to add up to 100%.
bSelf-reported.
Results of “Feed and Swaddle” Protocol According to Patient Characteristics.a
Abbreviations: CT, computed tomography; SD, standard deviation.
aDue to rounding, percentages may not always appear to add up to 100%.
bSelf-reported.
Discussion
We found that the “feed and swaddle” method is an effective alternative to sedation for pediatric craniofacial imaging. Hansen reported the greatest success with this method in patients <12 weeks of age. In our study, we were able to achieve successful scans in over 90% of patients with an average age of 24 weeks (Hansen, 2009). This is the first report of success in a large number of patients with a nonsedative protocol in this age-group. Additionally, we were able to implement this protocol to obtain a scan in a patient of 60 weeks of age. Due to the disproportionately large number of successful scans with the “feed and swaddle” method, we were unable to make useful statistical comparisons between the characteristics of the successful and unsuccessful patients. The protocol was overwhelmingly successful with both term and preterm infants, children with and without siblings, with families who had to travel long or short distances, and patients with or without craniosynostosis found on CT scan. These findings encourage the broad application of this protocol as there were no overt barriers to a successful scan, so long as the protocol was followed.
There is no consensus on the risks of anesthesia and sedation on the developing brain. However, the Food and Drug Administration recently announced changes to the labeling for sedatives and general anesthetics cautioning that these medicines “may negatively affect brain development in children younger than 3 years” (FDA approves label changes for use of general anesthetic and sedation drugs in young children, 2016).
The other benefits of avoiding sedation—cost, logistical and scheduling constraints, staffing, patient satisfaction, and emotional stress—were not directly measured in our study. The charges for sedation and anesthesia staff services for CT imaging at our institution is estimated to be US$800, including drug costs, anesthesia staff, and recovery room staffing. Anecdotally, families consistently expressed gratitude for avoiding sedation and have high satisfaction with the protocol.
Successful implementation of the “feed and swaddle” protocol requires collaboration between anesthesia, plastic surgery, neurosurgery, radiology, nursing, and child life clinicians. At OHSU, child life specialists were instrumental in optimizing adherence to the protocol by facilitating phone counseling and printed brochures as well as overseeing implementation of the protocol and addressing fear and anxiety on the day of the procedure. Their expertise proved invaluable in implementing this program, as has been noted elsewhere (McGee, 2003).
Other investigators have studied patient factors such as child temperament in relation to pain and sedation (Lochary et al., 1993; Lee and White-Traut, 1996; Voepel-Lewis et al., 2000). In particular, longer attention span and a less active child, as evaluated by a preprocedural questionnaire, have been associated with completing a successful nonsedated imaging study in children aged 4 to 7 years (Voepel-Lewis et al., 2000). For infants, there are no guidelines for identifying the patients who may not be successful with the “feed and swaddle” or other similar protocols.
The limitations of our study include small number of patients and patient selection. Craniofacial clinicians interviewed families to assess their ability to participate, introducing the possibility of selection bias. Unfortunately, information on how many patients were offered enrollment and how many declined was not recorded. The intentional selection bias for cooperative families who judged their infant likely to cooperate with the protocol likely resulted in the very high success rate observed in this study and represents a test of real-world feasibility in clinical practice. Any attempt at uniform adoption for all patients and families, however, would require a more formal intention to treat paradigm and collection of rigorous enrollment statistics. The high protocol success rate also precluded a statistical comparison of patient characteristics between successful and nonsuccessful infants. Finally, the demographics of our patients were largely white, non-Hispanic patients. This may limit external validity of the study in centers with different patient populations and socioeconomic challenges.
Conclusions
Sedation is routinely used to achieve motionless pediatric imaging but is also associated with possible neurodevelopmental risks and increased use of resources. The “feed and swaddle” protocol is an effective alternative to sedation for pediatric craniofacial imaging in patients up to the age of 60 weeks. Further research is needed to identify factors associated with successful implementation of this protocol.
Footnotes
Acknowledgment
The authors thank Shirley McCartney, PhD, for editorial assistance.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.