Abstract
Mandibular distraction osteogenesis (MDO) is increasingly used for neonates and infants with upper airway obstruction secondary to micrognathia. This systematic review was conducted to determine the effectiveness of MDO in the treatment of airway obstruction. The databases searched included PubMed, Embase, Scopus, and grey literature sources. The inclusion criteria were applied to identify studies in children with clinical evidence of micrognathia/Pierre Robin sequence (PRS) who had failed conservative treatments, including both syndromic and non-syndromic patients. Overall 66 studies were included in this review. Primary MDO for the relief of upper airway obstruction was found to be successful at preventing tracheostomy in 95% of cases. Syndromic patients were found to have a four times greater odds of failure compared to those with isolated PRS. The most common causes of failure were previously undiagnosed lower airway obstruction, central apnoea, undiagnosed neurological abnormalities, and the presence of additional cardiovascular co-morbidities. MDO was less effective (81% success rate) at facilitating decannulation of tracheostomy-dependent children ( P < 0.0001). Failure in these patients was most commonly due to severe preoperative gastro-oesophageal reflux disease, swallowing dysfunction, and tracheostomy-related complications. The failure rate was higher when MDO was performed at an age of ≥24 months. More studies are needed to evaluate the long-term implications of MDO on facial development and long-term complications.
Micrognathia is a congenital condition involving an abnormally small mandible. This condition tends to occur in conjunction with posterior tongue displacement (glossoptosis), which can lead to physical obstruction of the oropharyngeal and hypopharyngeal regions on inspiration. This upper airway obstruction may be life-threatening and may require urgent medical intervention.
In 1923, a French stomatologist was the first to describe the constellation of symptoms associated with upper airway obstruction in neonates now known as Pierre Robin sequence (PRS). This sequence is a craniofacial anomaly characterized by mandibular micrognathia (mandibular hypoplasia), glossoptosis, and in some cases a ‘U-shaped’ cleft palate. There is only limited epidemiological data, but the incidence has been reported to range from approximately 1 in every 8500 live births in Liverpool, UK to 1 in 14,000 live births in Denmark. The most recent study from Germany reports an incidence of approximately 1 in 8000 births. This variation in incidence is related in part to the inconsistent definition of PRS in the literature.
The diagnosis of patients with PRS is challenging due to the wide spectrum of PRS phenotypes, variation in degree of airway obstruction, feeding difficulties, and the need for treatment. This has led to some authors only characterizing those with airway obstruction needing treatment as having PRS ; others will include all patients with micrognathia and glossoptosis, or limit the PRS diagnosis to those with associated cleft palates. Although these clinical features are most commonly seen in isolation, they can also occur in association with other syndromes of the craniofacial skeleton; for example, Treacher Collins syndrome, Stickler syndrome, and Nager syndrome. Such co-occurrences further complicate the diagnosis. These syndromes differ in pathogenesis from isolated PRS, but all can have micrognathia with glossoptosis and hence airway obstruction. For simplicity, those without an associated syndrome are referred to in this study as having ‘isolated PRS’ and those with an associated syndrome are referred to as having ‘syndromic micrognathia’. The varying phenotypes and presumed causes of this anomaly make comparison of the myriad of protocols advocated for management difficult.
The most important consequences of micrognathia and PRS are the inability to effectively breathe or feed due to airway obstruction. The majority of children born with micrognathia or PRS have no respiratory distress. Those with mild symptoms of respiratory distress can often be treated conservatively with prone positioning or non-invasive techniques, such as a nasopharyngeal airway or the application of nasal continuous positive airway pressure (CPAP). The rate of success with the use of nasopharyngeal airways varies in the literature, ranging from 48% to 100%. A large case series study focusing on children with non-syndromic PRS demonstrated that less than 10% required a surgical intervention.
For neonates with severe respiratory distress, or those who fail initial conservative treatment, the airway dysfunction can be a life-threatening emergency. The nasopharyngeal airway and CPAP can only be tolerated for a limited period of time, and in some cases, children need to be intubated and ventilated to maintain adequate oxygenation. Children who require prolonged treatment with these measures may require a more definitive surgical intervention.
Several surgical treatments have been described for the treatment of the child with micrognathia. In 1946 Douglas described the use of tongue–lip adhesion (TLA) for the treatment of upper airway obstruction associated with micrognathia. This procedure involves surgically fusing the tongue to the anterior lower lip to hold the tongue in an anterior position. The adhesion is usually reversed with another surgical procedure at 9–12 months of age. However, the underlying cause of the obstruction is not fully addressed by TLA, and wound dehiscence and feeding difficulties are common, thus many centres have abandoned it as a viable treatment option.
Other surgical options described include mandibular traction and advancement appliances, and sub-periosteal release of the floor of the mouth musculature. However these procedures have not met with widespread success and have largely been abandoned by the larger centres.
Mandibular distraction osteogenesis (MDO)
Since the introduction of distraction osteogenesis for the craniofacial skeleton in the mid-1980s to early 1990s, it has been used to deal with various types of reconstructive dilemma. MDO for infants with micrognathia has been used for unilateral mandibular lengthening by distraction for cases of hemifacial microsomia and bilateral MDO for cases of Treacher Collins syndrome. Initially MDO was used to resolve upper airway obstruction and to facilitate the removal of a tracheostomy. Since then, it has been used increasingly as the primary surgical option for the management of neonates and infants with micrognathia or PRS with upper airway obstruction.
MDO relieves the airway obstruction by lengthening the mandible. This stretches the tongue attachments to the mandible (genioglossus muscle), which positions the tongue more anteriorly, relieving the glossoptosis. Most children with upper airway obstruction have demonstrated an improvement in their respiratory status within a few days of distraction. For those children who are intubated and mechanically ventilated, this may mean extubation and transfer to a regular hospital ward.
Several case series have demonstrated the effectiveness of MDO in alleviating upper airway obstruction in neonates, infants, and older children with PRS. Most patients were able to avoid tracheostomies, and those who already had tracheostomies could be decannulated. A systematic review performed in 2008 evaluated the effectiveness of MDO in several clinical applications. The review evaluated 178 studies including 1185 patients. Success in preventing tracheostomies was achieved in 91.3% of patients. However, the authors of that review searched only the PubMed database on the applications of unilateral and bilateral mandibular distraction in both children and adults. Limiting the search to a single database is a significant methodology limitation of that review. In addition the study also included all possible causes of micrognathia, including temporomandibular joint (TMJ) ankylosis, hemifacial microsomia, and syndromic micrognathia, which have different aetiologies to isolated PRS. No comparative subgroup analyses were performed to differentiate between these groups. Furthermore, the authors did not evaluate any long-term outcomes in children and did not discuss reasons for failure of distraction.
This current review was performed with the aim of extending the search across multiple databases to include the current available evidence for the effectiveness of mandibular distraction for the treatment of upper airway obstruction in children with micrognathia. This review also reports reasons for failure and compares outcomes between isolated PRS/micrognathia patients and syndromic micrognathia patients. A further aim is to determine the effects of mandibular distraction on the other complications of micrognathia, including feeding and weight gain, gastro-oesophageal reflux, and facial development. These outcomes will be presented in subsequent papers.
Methods
A research protocol was written and peer-reviewed prior to undertaking this systematic review. The PICO criteria used for this review are listed in Table 1 . All of the studies in this review included children with clinical evidence of micrognathia who underwent bilateral MDO and also children who had initially undergone conservative treatment options. This review considered reasons for failure and for the consideration of tracheostomy. Syndromic and non-syndromic children were included, but some particular conditions such as bilateral TMJ ankylosis, hemifacial microsomia, and other conditions that may contribute to the airway obstruction for reasons other than the micrognathia alone, were excluded. In addition, children with known lower airway abnormalities prior to treatment were also excluded.
| Participants | • Male and female children from birth with clinical evidence of micrognathia |
| • Clinical evidence of upper airway obstruction, who failed conservative treatments | |
| • Syndromic and non-syndromic children | |
| • Bilateral mandibular distraction | |
| • Minimum of 1 year follow-up | |
| Exclusion: | |
| • Children who underwent unilateral distraction | |
| • Children with known preoperative central apnoea/lower airway abnormalities | |
| • TMJ ankylosis/hemifacial microsomia or other mandibular condition leading to airway obstruction | |
| Intervention | Bilateral mandibular distraction osteogenesis |
| Comparator | Tracheostomy |
| Outcomes | Airway outcomes |
| • Primary mandibular distraction osteogenesis | |
| • Decannulation of tracheostomy-dependent patients | |
| Long-term outcomes | |
The types of studies considered included both experimental and epidemiological study designs, including randomized control trials, quasi-experimental studies, prospective and retrospective cohort studies, and case–control studies. The review also considered case series and case reports where higher levels of evidence were not found.
The search strategy included both published and unpublished studies in English, covering the period 1990 to November 2013. The databases searched included PubMed, CINAHL, EMBASE, SCOPUS, Web of Knowledge, and other grey literature databases such as Scirus and MedNar. An example of the strategy used to search through PubMed is given in Fig. 1 .
Papers selected for retrieval were assessed by two independent reviewers for methodological validity and were critically appraised using the standardized critical appraisal instruments of the Joanna Briggs Institute Meta Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI). Any disagreements that arose between the reviewers were resolved through discussion, or with a third reviewer. Data were extracted from the studies included and entered into study-specific Microsoft Excel tables. A second reviewer checked data extraction.
The data items extracted differed between outcomes. Where available, individual patient data were collected; where not available, the authors were contacted for clarification. For all studies, demographic data and patient characteristics were collected for all patients, including syndromic status. Follow-up periods were also recorded for the long-term outcomes analysis. Other data parameters recorded for each outcome and the definitions used for success and failure of treatment are presented in Table 2 .
| Outcome | Success | Failure |
|---|---|---|
| Primary mandibular distraction osteogenesis | Relief of upper airway obstruction | Failure: Needing further airway adjuncts due to persistent airway obstruction. The additional interventions needed were also recorded Polysomnography measures: obstructive apnoea/hypopnoea index (OAHI) Reasons for failure |
| Tracheostomy decannulation | Decannulation was achieved after MDO | Failure: Persistent tracheostomy despite MDO Reasons for failure |
The quantitative data collected and effect sizes are expressed as odd ratios (OR) with the 95% confidence intervals (95% CI). Subgroup analyses were performed where possible to compare syndromic and non-syndromic children with micrognathia and age at time of surgery. Where statistical pooling was not possible, the findings are presented in narrative form, including tables and figures to aid in data presentation.
Methods
A research protocol was written and peer-reviewed prior to undertaking this systematic review. The PICO criteria used for this review are listed in Table 1 . All of the studies in this review included children with clinical evidence of micrognathia who underwent bilateral MDO and also children who had initially undergone conservative treatment options. This review considered reasons for failure and for the consideration of tracheostomy. Syndromic and non-syndromic children were included, but some particular conditions such as bilateral TMJ ankylosis, hemifacial microsomia, and other conditions that may contribute to the airway obstruction for reasons other than the micrognathia alone, were excluded. In addition, children with known lower airway abnormalities prior to treatment were also excluded.
| Participants | • Male and female children from birth with clinical evidence of micrognathia |
| • Clinical evidence of upper airway obstruction, who failed conservative treatments | |
| • Syndromic and non-syndromic children | |
| • Bilateral mandibular distraction | |
| • Minimum of 1 year follow-up | |
| Exclusion: | |
| • Children who underwent unilateral distraction | |
| • Children with known preoperative central apnoea/lower airway abnormalities | |
| • TMJ ankylosis/hemifacial microsomia or other mandibular condition leading to airway obstruction | |
| Intervention | Bilateral mandibular distraction osteogenesis |
| Comparator | Tracheostomy |
| Outcomes | Airway outcomes |
| • Primary mandibular distraction osteogenesis | |
| • Decannulation of tracheostomy-dependent patients | |
| Long-term outcomes | |
The types of studies considered included both experimental and epidemiological study designs, including randomized control trials, quasi-experimental studies, prospective and retrospective cohort studies, and case–control studies. The review also considered case series and case reports where higher levels of evidence were not found.
The search strategy included both published and unpublished studies in English, covering the period 1990 to November 2013. The databases searched included PubMed, CINAHL, EMBASE, SCOPUS, Web of Knowledge, and other grey literature databases such as Scirus and MedNar. An example of the strategy used to search through PubMed is given in Fig. 1 .
Papers selected for retrieval were assessed by two independent reviewers for methodological validity and were critically appraised using the standardized critical appraisal instruments of the Joanna Briggs Institute Meta Analysis of Statistics Assessment and Review Instrument (JBI-MAStARI). Any disagreements that arose between the reviewers were resolved through discussion, or with a third reviewer. Data were extracted from the studies included and entered into study-specific Microsoft Excel tables. A second reviewer checked data extraction.
The data items extracted differed between outcomes. Where available, individual patient data were collected; where not available, the authors were contacted for clarification. For all studies, demographic data and patient characteristics were collected for all patients, including syndromic status. Follow-up periods were also recorded for the long-term outcomes analysis. Other data parameters recorded for each outcome and the definitions used for success and failure of treatment are presented in Table 2 .
| Outcome | Success | Failure |
|---|---|---|
| Primary mandibular distraction osteogenesis | Relief of upper airway obstruction | Failure: Needing further airway adjuncts due to persistent airway obstruction. The additional interventions needed were also recorded Polysomnography measures: obstructive apnoea/hypopnoea index (OAHI) Reasons for failure |
| Tracheostomy decannulation | Decannulation was achieved after MDO | Failure: Persistent tracheostomy despite MDO Reasons for failure |
The quantitative data collected and effect sizes are expressed as odd ratios (OR) with the 95% confidence intervals (95% CI). Subgroup analyses were performed where possible to compare syndromic and non-syndromic children with micrognathia and age at time of surgery. Where statistical pooling was not possible, the findings are presented in narrative form, including tables and figures to aid in data presentation.
Results
The search identified a total of 4815 studies. Out of these, a total 801 studies were retrieved based on title. After removal of duplicates, studies not in English, and studies outside the date criteria, only 382 studies remained. The abstracts were then reviewed to determine their relevance to the review question and objectives. During this process, 258 studies were excluded, leaving 124 studies that were retrieved for full-text examination ( Fig. 2 ).
Following the review of the full text, an additional 38 studies were excluded as they did not meet the inclusion criteria. Eighty-six studies were subjected to critical appraisal resulting in 66 studies included in the final analysis. The analysis included other outcomes not discussed in this paper; these will be presented in subsequent publications.
Overall, for the primary mandibular distraction outcome, 51 studies were included ( Table 3 ). The most common reason for exclusion was the potential risk of overlapping patients. In some of the studies with overlapping patients, different outcomes were reported in the different study reports. In these cases, all studies were included and the specific outcome data extracted from the individual studies. Within the subgroup analysis comparing syndromic and non-syndromic children, some studies were excluded due to inadequate distinction of which patients were syndromic and which had isolated PRS. Similarly, for the age-based subgroup analysis, studies were excluded due to limited age-based data to be able to make accurate comparisons.
| Article | Study design | Overall analysis | Syndromic vs. non-syndromic | Age-based |
|---|---|---|---|---|
| Al-Samkari 2010 | RR | Excluded – overlap | Included | Excluded – no age data |
| Andrews 2013 | RR | Included | Excluded – inadequate distinction | Included |
| Breugem 2012 | RR | Included | Included | Included |
| Brevi 2006 | CR | Excluded – overlap | Excluded – overlap | Excluded – overlap |
| Burstein 2005 * | RR | Included | Included | Excluded – no age data |
| Carls 1998 | RR | Included | Included | Included |
| Chigurupati 2004 | RR | Included | Included | Included |
| Chowchuen 2011 | RR | Included | Included | Included |
| Dauria 2008 | RR | Included | Included | Included |
| Denny 2005 | RR | Included | Included | Included |
| Genecov 2009 | RR | Included | Included | Included |
| Gifford 2008 * | RR | Included | Included | Included |
| Gözü 2010 | RR | Included | Included | Excluded – no age data |
| Griffiths 2013 * | CR | Included | Included | Included |
| Hammoudeh 2012 * | RR | Included | Included | Included |
| Handler 2009 | CR | Included | Included | Included |
| Hong 2012 | RR | Included | Included | Included |
| Hong 2012 | RR | Included | Included | Included |
| Howlett 1999 | CR | Included | Included | Included |
| Izadi 2003 | RR | Excluded – overlap | Included | Excluded – overlap |
| Judge 1999 | CR | Included | Included | Included |
| Kolstad 2011 | RR | Included | Included | Included |
| Lee 2009 | RR | Included | Included | Included |
| Lin 2006 * | RR | Included | Included | Included |
| Looby 2009 * | RR | Included | Included | Included |
| Mandell 2004 | RR | Included | Included | Included |
| Miller 2007 * | RR | Included | Included | Excluded – no age data |
| Miloro 2010 | RR | Included | Excluded – inadequate distinction | Excluded – no age data |
| Mitsukawa 2007 * | RR | Included | Included | Included |
| Monasterio 2002 * | RR | Excluded – overlap | Excluded – overlap | Excluded – no age data |
| Monasterio 2004 | RR | Included | Included | Included |
| Morovic 2000 | PR | Included | Included | Included |
| Mudd 2012 | RR | Included | Excluded – inadequate distinction | Included |
| Murage 2013 * | RR | Included | Included | Included |
| Olson 2011 | RR | Included | Included | Excluded – no age data |
| Papoff 2013 | RR | Included | Included | Included |
| Perlyn 2002 | RR | Included | Included | Included |
| Rachmiel 2012 | RR | Included | Excluded – inadequate distinction | Included |
| Sadakah 2009 * | RR | Included | Included | Included |
| Schaefer 2004 | RR | Included | Included | Included |
| Scott 2011 | RR | Excluded – overlap | Included | Included |
| Sesenna 2012 | RR | Included | Included | Included |
| Sidman 2001 | PR | Excluded – overlap | Included | Included |
| Smith 2006 | RR | Included | Included | Included |
| Sorin 2004 | RR | Included | Included | Included |
| Spring 2006 | RR | Excluded – overlap | Excluded – overlap | Included |
| Taub 2012 | CR | Included | Included | Included |
| Tibesar 2006 | CR | Included | Included | Included |
| Tibesar 2010 | RR | Included | Excluded – inadequate distinction | Excluded – no age data |
| Wittenborn 2004 | RR | Included | Excluded – inadequate distinction | Included |
| Zenha 2012 | CR | Included | Included | Included |
| 44 included | 42 included | 41 included |
* Studies that included polysomnography results pre- and post-MDO.
Primary mandibular distraction osteogenesis
The primary MDO analysis included patients who underwent mandibular distraction as the primary surgical intervention after failing conservative therapy for upper airway obstruction. All of these patients had failed non-surgical therapy and were being considered for a tracheostomy. A successful outcome was defined as prevention of a tracheostomy and the relief of upper airway obstruction. There were a variety of ways in which this outcome was reported in these studies. Most reported only subjective improvements in obstructive symptoms, including reduction in noisy breathing and in clinically obvious apnoeic episodes. Other studies reported objective improvements, such as the ability to successfully extubate the patient who was otherwise intubated because of inability to self-ventilate, or to maintain normal oxygen saturation on room air as measured by pulse oximetry. Eleven studies included polysomnography results pre- and post-MDO (marked with an asterisk (*) in Table 3 ). Failure was defined as requiring a tracheostomy despite mandibular distraction due to persistent airway obstruction.
This analysis was based on data obtained from 44 papers involving 490 patients. The mean age was 10.4 months (range 5 days to 8 years). Amongst these patients, 468 had a successful outcome, while 22 required a tracheostomy. This equates to an overall success rate of 95.5% for mandibular distraction preventing tracheostomy in the studies included ( Table 4 ). Amongst the successful outcomes, two patients required home oxygen in the short term, but avoided any further surgical intervention and were able to be discharged home. One patient required nocturnal CPAP for 3 years after distraction, but did not require any further surgical intervention.
