Abstract
Objectives
To report survey results from American Cleft Palate–Craniofacial Association members on the practice patterns of airway obstruction management in patients with Pierre Robin sequence.
Design
A 10-question online survey was sent and the data were reviewed.
Setting
Online survey of members of the American Cleft Palate–Craniofacial Association.
Patients
Surveys assessed management patterns of patients with Pierre Robin sequence whom a surgeon member of the American Cleft Palate–Craniofacial Association treated for airway obstruction.
Interventions
The survey comprised data on management strategies for airway obstruction in Pierre Robin sequence, including tracheostomy, tongue-lip adhesion, mandibular distraction, and treatments that falls in the “other” category.
Results
A total of 87 American Cleft Palate–Craniofacial Association members completed the survey. Respondents' results were analyzed as a whole and by individual subspecialty: plastic surgery (n = 33), oromaxillofacial surgery (n = 21), and otolaryngology (n = 29). Although most of the surgeons were trained to manage airway obstruction in Pierre Robin sequence patients using tracheostomy (47%, n = 39) and tongue-lip adhesion (31%, n = 26), 48% reported a current preference for mandibular distraction (n = 40). Of surgeons who preferred to manage Pierre Robin sequence with tongue-lip adhesion (n = 23), 65% were trained to do so (n = 15). Surgeons preferring mandibular distraction (n = 40) and tracheostomy (n = 14) more often reported they were trained to manage Pierre Robin sequence with tracheostomy.
Conclusions
Currently there are various practice patterns for the management of airway obstruction in Pierre Robin sequence. Training habits and subspecialty category may influence a surgeon's preference in patients who fail conservative therapy. Treatment guidelines are lacking and may require significant collaboration among centers and subspecialties to develop a more standardized approach to a challenging clinical entity.
In 1923, Pierre Robin first described a set of findings including glossoptosis and micrognathia that led to breathing difficulty. By 1934 he noted the association of a cleft palate (Robin, 1934). Although Pierre Robin sequence (PRS) has evolved to encompass a variety of definitions, the standardized definition consisting of micrognathia, glossoptosis, and airway compromise is most widely accepted and will be adhered to in this discussion (Breugem, 2009). Pierre Robin sequence occurs in approximately 1 per 8500 to 1 per 14,000 births (Bush and Williams, 1984; Printzlau and Andersen, 2004). Mandibular hypoplasia in utero displaces the tongue posteriorly and superiorly, preventing palatal shelf fusion and causing the classic micrognathia and glossoptosis. Glossoptosis contributes to varying degrees of tongue-base obstruction, sleep apnea, and respiratory distress (Meyer et al., 2008). Additional factors also contribute to dysphagia, including tongue prolapse into a cleft palate, lack of voluntary control of tongue musculature, and a negative pressure pull of the tongue into the hypopharynx (Evans et al., 2011).
Currently there are no official practice guidelines for the surgical management of airway obstruction in patients with PRS who fail conservative treatment (Tomaski et al., 1995). The traditional approach is tracheostomy, which remains a first-line surgical therapy for some surgeons (Genecov et al., 2009). Most craniofacial centers currently use tongue-lip adhesion (TLA) and/or mandibular advancement procedures (Denny and Amm, 2005; Genecov et al., 2009). The literature is unclear as to which surgical airway procedure for PRS patients is most effective. It is even unclear if there is a “one surgery fits all” type of approach that is superior to a more adaptable and patient-dependent approach (Mackay, 2011). Studies have identified patient factors that may help guide management decisions (Kirschner et al., 2003; Schaefer et al., 2004; Dauria and Marsh, 2008), but reliable, high-evidence studies are difficult considering the low incidence of pathology and the age of the patient population (Elluru, 2005). Each procedure has significant potential complications that must be considered. Other factors such as surgeon training and experience may also play key roles in the decision process (Mackay, 2011). The purpose of this study is to identify the current PRS management strategies and practice patterns of American Cleft Palate–Craniofacial Association (ACPA) members from various training backgrounds who surgically treat PRS airway obstruction.
Methods
A 10-question online survey (Fig. 1) was sent to the ACPA membership, and members were invited to participate. A total of 1109 electronic mail invitations were sent to ACPA members from all disciplines asking surgeons to complete a survey on SurveyMonkey (http://www.surveymonkey.com). The survey was sent only once in September 2010 to avoid repeat respondents. Each member participant reported his or her specialty and perceptions regarding the treatment and surgical management of PRS patients within his or her own practice. Results from 83 respondents within one of three subspecialties that included plastic surgery (n = 33), oromaxillofacial surgery (n = 21), and otolaryngology (n = 29) were analyzed. Four respondents who categorized their specialty as “other” were not included in specialty-specific evaluations because the category would have been too small for adequate analysis. Data were compiled by the SurveyMonkey program (2010) and analyzed.
American Cleft Palate–Craniofacial Association members survey on the management of airway obstruction in patients with Pierre Robin sequence.
Results
A summary of the survey responses by question can be found in Table 1. Surgeons most frequently reported that they were trained to manage airway obstruction in PRS patients using tracheostomy (47%, n = 39) and TLA (31 %, n = 26). When asked which primary surgical treatment they preferred for managing airway obstruction among PRS patients who failed to respond to nonoperative treatment, surgeons most often reported a preference for mandibular distraction (MD) (48%, n = 40) and TLA (28%, n = 23). The majority of plastic surgeons (67%, n = 22) and otolaryngologists (76%, n = 22) reported they see on average six or more infants with PRS in 1 year; whereas, the majority of oromaxillofacial surgeons (57%, n = 12) reported seeing five or fewer in 1 year (Fig. 2). The majority estimated that between 0% and 50% of their PRS patients are syndromic (92%, = 76), that 0% to 40% of PRS infants require surgery for airway obstruction (82%, n = 68), and that at the time of surgical treatment their PRS patients are ≤1 month of age (61%, n = 51) (Table 1).
Average number of new Pierre Robin infants seen per year, percentage by subspecialty.
Summary of Survey Responses Regarding the Management of Airway Obstruction in PRS Patients (n = 83)*
PRS = Pierre Robin syndrome; CI = confidence interval; TLA = tongue-lip adhesion.
Survey responses were also summarized by subspecialty (Figs. 1 through 5). Plastic surgeons most often reported they were trained to manage airway obstruction in PRS patients with TLA (48%, n = 16). The majority of oromaxillofacial (57%, n= 12) and otolaryngology (62%, n = 18) surgeons reported they were trained to treat PRS patients for airway obstruction using tracheostomy. Yet, training did not seem to be reflected in surgeons' reported preferences for primary management of airway obstruction for PRS patients for any of the subspecialties except plastic surgery. In accord with their training, plastic surgeons reported they prefer using TLA (42%, n = 14), followed by MD (30%, n = 10). Oromaxillofacial surgeons more frequently reported they preferred to manage airway obstruction in PRS patients using MD (43%, n = 9) and TLA (33%, n = 7). Approximately 73% (n = 21) of otolaryngology surgeons reported they preferred MD to treat PRS patients.
A review of the distribution among preferences for the primary treatment of airway obstruction in PRS patients by training of surgeons (Figs. 3 to 4 and 6) reveals that training may impact the preferred primary management of PRS patients for airway obstruction. Survey respondents reported that the most common or preferred procedure they use to treat their PRS patients was MD (48 %, n = 40), TLA (28%, n = 23), tracheostomy (17%, n= 14), and other (7%, n = 6). The majority (65%) of those who prefer to use TLA (n = 23) reported they were trained to use TLA (n = 15). Sixty-four percent of surgeons who preferred to use tracheostomy (n = 14) and 55% of those who preferred MD (n = 40) reported they were trained to use tracheostomy.
Percentage of subspecialists trained to perform TLA, MD, tracheostomy, or other primary surgical technique.
Percentage of subspecialists currently using TLA, MD, tracheostomy, or other primary surgical technique.
Percentage of infants with Pierre Robin sequence who require airway surgery, percentage by subspecialty.
Percentage of those primarily practicing TLA, MD, tracheostomy, or other procedures who were trained to use these respective techniques.
Discussion
In the absence of PRS practice guidelines, there remains wide variation in practice patterns and surgeon preference in the management of airway obstruction. Reasons that guidelines are lacking may include the highly variable phenotype of PRS, the persistent discrepancy in the diagnosis of PRS among providers, and difficulty in predicting which children will resolve their obstruction with conservative measures. Also, there is a lack of compelling comparative studies due to the age of the patient population and scarcity of this disease entity (Elluru, 2005). In the present era without PRS guidelines to direct more consistent training, patient care, and outcome measures, it is helpful to know current practice tendencies. Numerous surgical options are used to overcome the anatomical changes in PRS. Although the classic method is tracheostomy, with the exception of perhaps the most severe patients, it has now become a last resort for many surgeons for the treatment of PRS (Meyer et al., 2008). Currently, the most common alternatives to tracheostomy are TLA and MD, both of which have been shown to be effective in various studies (Parsons, 1982; McCarthy et al., 1992; Sidman et al., 2001; Kirschner et al., 2003; Denny et al., 2004; Denny and Amm, 2005; Meyer et al., 2008; Genecov et al., 2009). A number of authors have proposed management algorithms to follow when treating complicated PRS patients who fail conservative treatment. Some choose between MD and tracheostomy based on the patient's clinical status, need for long-term airway support, and gestational age (Dauria and Marsh, 2008). Mandibular distraction is recommended for PRS neonates with no compounding upper airway, cardiac, or pulmonary pathology contributing to the respiratory distress who were born after 38 weeks' gestation; if these criteria are not met, tracheostomy is performed as the primary surgical procedure. The small number of patients who qualify to receive MD seem to have favorable outcomes and can avoid tracheostomy (Dauria and Marsh, 2008). Another algorithm stratifies PRS patients into three subdivisions based on endoscopic findings: no visible obstruction, glottis/infraglottic obstruction, and tongue-base obstruction. Children with no visible obstruction are referred for a workup of a likely central nervous system or pulmonary disorder. Neonates with glottis/infraglottic obstruction are treated with a tracheostomy. The group with tongue-base obstruction that fails conservative treatment is initially treated with a TLA. If TLA is unsuccessful by the time the infant is 6 to 7 months old, repeat endoscopy and MD are pursued. If MD fails, a tracheostomy is performed—ultimately subjecting some patients to three or more airway surgeries (Schaefer et al., 2004). The Children's Hospital in Philadelphia follows a similar algorithm in PRS patients with no laryngotracheal abnormality by using TLA as the primary procedure, trying MD upon TLA failure, and resorting to tracheostomy if necessary (Kirschner et al., 2003).
Our survey shows that surgeons with a preference for TLA were most likely taught to perform TLA at their training institution. Similarly, the majority who prefer tracheostomy were more commonly taught in this manner (Fig. 6). Considering the more recent discovery and use of MD, most surgeons who prefer MD were not taught this procedure in training (the majority were taught to perform tracheostomy). Despite less than 20% of all surgeons receiving training to perform MD as the primary surgical technique, 30% of plastic surgeons, 43% of oromaxillofacial surgeons, and 73% of otolaryngologists preferred MD in these instances. In this survey, MD is the most likely primary surgery to be performed once conservative treatment fails. When surgeons use TLA as the initial surgery and a secondary technique is necessary, the majority will choose MD (Table 1).
Complications of tracheostomy, TLA, and MD are also important to consider when choosing a procedure. Common tracheostomy complications include suprastomal granulation and collapse, fistulas, tracheal stenosis, tube obstruction, accidental decannulation, creation of false passages, cellulitis, neck scarring, and loss of airway (Carr et al., 2001; Perdra et al., 2004). Up to 60% of infants undergoing tracheostomy may have some type of complication (Perdra et al., 2004). Infants undergoing TLA may have dehiscence, infection, submaxillary duct obstruction, lip scarring, postoperative obstructive sleep apnea, severe dysphagia requiring prolonged gastrostomy tube usage, and growth retardation with a high rate of secondary procedures performed (Hoffman, 2003; Denny et al., 2004). The complication rate has been reported from 17% to 43% (Hoffman, 2003; Kirschner et al., 2003). Complications of MD include wound infections, facial cellulitis, temporary paresthesia, facial nerve injury, scarring, cheek abscess, open bite deformity, tooth bud injury, jaw deformity, and dentigerous cyst formation (Hong, 2011). The most common complication reported by some is permanent dentition loss or malformation, at a rate of 21% (Scott et al, 2011). Recurrence of airway obstruction or feeding difficulties can occur with any of the three techniques.
This survey queried the outcomes expected by surgeons when performing respective procedures (Figs. 7 and 8). Surgeons who preferred tracheostomy most often perceived higher success rates in treating airway obstruction but poorer success rates in preventing gastrostomy tube placement than did those who prefer TLA and MD. It is unclear whether surgeons in various subspecialties may be trained to value differently the risks versus benefits in the management of PRS, but this may help explain phenomena such as the tendency to perform a tracheostomy to ensure an adequate airway despite a higher perceived risk of gastrostomy tube to the patient. It is interesting that other surgeons feel more comfortable performing an initial procedure with a lower perceived success rate (TLA or MD), but these surgeons also report a lower rate of gastrostomy tube placement.
Perceived success rate of primary surgery choice.
Perceived rate of gastrostomy tube following primary procedure.
The ideal management approach to PRS should encompass a variety of cost-effective treatment options that give the patient an opportunity for timely improvement of airway obstruction and dysphagia with the lowest risk of a decreased quality of life (Kohan et al., 2010). It has been established that in the majority of cases this can be accomplished with conservative measures alone (e.g., nasal trumpet) (Kirschner et al., 2003). The majority of our respondents estimated that only 0% to 20% of their patients require airway surgery (Fig. 5). In the patients who require surgery, it is necessary to evaluate each on an individual basis because those affected by PRS constitute not a homogeneous but a heterogeneous group inclusive of many subsets of patients. Secondary to causal, pathogenetic, and phenotypic differences, Robin complexes of various types can occur (Cohen, 1999). Of PRS cases, 20% to 30% may be associated with syndromes, but most patients have isolated PRS (Meyer et al., 2008; Bütow et al., 2009). In our survey the majority estimated that between 0% and 50% of their patients have syndromic PRS (92%, n = 76) (Table 1). This has important implications in growth expectations following treatment, given that it seems children with syndromic PRS are less likely to catch up and follow a normal growth curve than those with isolated PRS (Rogers et al., 2009; Mackay, 2011). In addition to different growth expectations, beneficial treatment pathways are likely to differ between these groups. If the condition of a PRS patient with upper airway obstruction is complicated by a lower airway glottic or subglottic obstruction, tracheostomy may be the most efficient and well-tolerated procedure to correct all abnormalities despite the side effects. In a patient with isolated PRS, MD or TLA may be the most effective and well-tolerated procedure depending on the experience of the surgeon. Because a variety of procedures may be appropriate for the treatment of PRS patients, it may be important that training programs allow surgeons to become comfortable with each option. Otherwise, surgeons should be involved in a craniofacial team that can comprehensively provide individualized options.
A limitation of our survey-based study is the inevitability of selection bias. Although 1109 e-mail accounts were contacted initially, some ACPA members may have had multiple accounts included on this list. It is also unclear how many of the 1109 were surgeons; thus, it is not possible to calculate an accurate response rate from the surgeons asked to participate. The 83 members of the ACPA who responded were more likely to do so secondary to strong opinions or expertise regarding the topic. In addition, self-reporting of one's own surgical successes and complications may naturally tend toward a favorable perception. Although there was no primary gain from participation in our anonymous survey, secondary gain from the unconscious desire to correlate a specialty with positive surgical outcomes could have influenced these reports. The sample size of 83 respondents is relatively small. Any retrospective survey is limited to some extent by recall bias, although we assume that the period of recall is brief in most instances because the respondents are actively practicing. The recall of training preference should have been straightforward regardless of elapsed time.
In the absence of practice guidelines for airway management of patients with PRS, it is helpful to understand the current preferences of practicing surgeons. If guidelines can be established, they will be based on the most effective treatments and algorithms demonstrated by the majority of surgeons. Our survey shows that tracheostomy, TLA, and MD are all used in treatment, but training habits and subspecialty category seem to currently have a strong influence on a surgeon's preference. Surgeons recognize the difference in the likelihood of complications between these procedures which may impact preference as well. Further collaborative outcome studies are needed to examine the usefulness of these techniques. Because there is likely a place for all three procedures in the care of PRS patients, it is important that these techniques are shared with all subspecialties involved in treatment and implemented in the respective training programs. Care may improve if evidence from practice habits is widely distributed and if the implementation of dedicated craniofacial teams can give each patient with PRS individualized and comprehensive treatment.