Perioperative Management of Obstructive Sleep Apnea in Patients With Syndromic Craniosynostosis Undergoing LeFort III Osteotomy With Distraction: A Case Series

Introduction

The syndromic craniosynostoses (i.e., Crouzon, Apert, and Pfeiffer syndromes) represent a relatively rare grouping of genetic mutations. Gain of function mutations in the Fibroblast growth factor receptor (FGFR) gene family are directly associated with the bony abnormalities which characterize these genetic syndromes. ¹ These patients are often affected by bicoronal craniosynostosis, anterior open bite malocclusions, along with their own characteristic abnormalities. ² Respiratory compromise often develops due to narrowing of the nasopharyngeal airway caused by retrusion of the midface. Obstructive sleep apnea (OSA) is a particularly problematic complication of progressive midface retrusion occuring in a approximately 50% of these patients. ³ Nighttime apnea may be life-threatening, with severe OSA occurring most commonly in patients with Pfeiffer syndrome.^4,5 In contrast to the average adult population, continuous positive airway pressure (CPAP) is often insufficient for definitive treatment of OSA in this pediatric population. The characteristic facial topography of patients with syndromic craniosynostosis complicates mask fit, often compromising patient comfort and treatment efficacy. ⁶ Additionally, patients in this population who are well managed through CPAP initially are unlikely to maintain appropriate mask fit as midface hypoplasia worsens with growth. Therefore, operative management is typically required. ⁷

Operative midface advancement with distraction osteogenesis has become the “Gold Standard” in managing OSA in this patient population and is supported by a greater body of evidence compared to less invasive methods, including CPAP. ⁸ However, typical distraction protocols do not begin until several days after LeFort III procedure and hardware placement and several days of distraction may be required to achieve midface advancement sufficient to ameliorate sleep-disordered breathing. During this time, the presence of distraction hardware and postoperative midface instability limit options for nighttime airway management should respiratory compromise occur.

In a recent review, Garcia-Marcinkiewicz and Stricker identified the limitations of management options caused by hardware placement and provided several strategies for minimizing the risk of postoperative respiratory compromise (e.g., intraoperative dexamethasone). ⁹ Nevertheless, the authors did not provide specific recommendations for immediate management of patients for whom sufficient distraction cannot be achieved for an extended time postoperatively. ⁹ In a case series including seven patients undergoing LeFort III procedures for craniosynostosis, tracheostomies were conducted to ensure proper respiration postoperatively. ¹⁰ However, given the relative rarity of severe respiratory compromise following this operation, ¹¹ prophylactic tracheostomy may not be appropriate for all patients. Unfortunately, literature detailing minimally invasive respiratory management options in the immediate postoperative period prior to initiation of distraction protocol is conspicuously absent. ⁹ The need for clear postoperative management protocols is amplified by continued surgical revision, which is often required in this patient population.

Within prior literature, there is no agreed upon protocol for the noninvasive management of postoperative respiratory distress in this unique patient population. The present report aims to present such a protocol, with specific attention paid to the external halo distraction device, which complicates oxygen delivery options.

Methods

Given the relative rarity of the syndromic craniosynostoses, the time horizon was set such that all patients who underwent elective LeFort III osteotomy with a KLS-Martin halo distractor placement within roughly one calendar year to allow for sufficient patient accrual. All procedures were executed by the same primary surgeon at a single publicly funded tertiary care center. Four patients with syndromic craniosynostosis, ranging from 10 to 19 years of age underwent the procedure within the study period. Three of the four patients were identified as having significant obstructive sleep apnea as a primary indication for operative management. Of those with OSA, preoperative apnea-hypopnea index (AHI) ranged from approximately 12.2–28.5 (median 28.3). All patients were initially given a diagnosis of Crouzon syndrome. However, one patient was reclassified by mutation analysis as having Pfeiffer Syndrome before commencement of the study period.

Bicoronal exposure and LeFort III osteotomy and halo distractor placement were performed in the standard fashion in all four patients. All patients received scheduled dexamethasone postoperatively and were admitted to the pediatric intensive care unit (PICU) for close respiratory monitoring. Distraction of 1 mm/day was begun on postoperative day four or five. Two view plain films were obtained prior to initiation of distraction and weekly thereafter during the distraction period. Patients 2, 3, and 4 experienced no significant postoperative respiratory compromise. For the purpose of this communication, patient 1 will be highlighted due to her significant postoperative respiratory compromise. A brief overview of patients 2–4 is also included, as shown in Supplemental Table 1.

Case Reports

Patient 1

This patient was a 10-year-old female with Pfeiffer syndrome. At birth, the patient experienced difficulty breathing during feeds requiring admission to the neonatal intensive care unit (NICU). Her mother passed away in her 20s due to sleep apnea, likely sequelae of syndromic craniosynostosis. Secondary to her syndromic diagnosis, she was affected by bilateral hearing loss, sleep-disordered breathing, juvenile scoliosis, craniofacial abnormalities, craniosynostosis, and abnormalities of the bilateral thumbs.

At presentation, she had undergone multiple lifetime operations. Within the first year of life, she presented with apex ulnar angulation of the proximal phalanx of her thumbs bilaterally requiring surgical correction. Following this first operation, she required several days of supplemental oxygen to avoid desaturation. At age two, she required cranial vault remodeling for craniosynostosis. At age three, she underwent evaluation under anesthesia to assess potential hearing loss due to Pfeiffer’s syndrome. While recovering from anesthesia, she was noted to have substernal and intercostal retractions with upper airway obstruction requiring oxygen by facemask to maintain saturation.

At 5 years old, the patient was diagnosed with obstructive sleep apnea by polysomnography at which time she had an apnea-hypopnea index (AHI) of 37.6. She was titrated on BiPAP for nightly management. She was additionally found to have adenoidal and tonsillar hypertrophy contributing to sleep-disordered breathing. She subsequently underwent uncomplicated tonsillectomy & adenoidectomy which improved nighttime respiratory status. However, repeat polysomnography at age 7 revealed continued, albeit improved, OSA with an AHI of 12.2. By age 8, she required revision of her prior adenoidectomy due to regrowth. At 9 years old she presented to plastic surgery with severe midface hypoplasia, exorbitism, and class III malocclusion due to Pfeiffer syndrome. At that time, LeFort III midface advancement with external distraction osteogenesis was performed.

She was extubated postoperatively and admitted to the pediatric intensive care unit (PICU). While admitted to the PICU, she experienced multiple nightly obstructive desaturations with PaO₂ in the 50s. Multiple attempts at repositioning the patient during sleep failed to improve oxygenation and external distraction hardware prevented the utilization of continuous positive airway pressure (CPAP). Ultimately, nighttime oxygenation status was maintained through continuous monitoring by the care team who awakened the patient multiple times throughout the night during apneic episodes. Due to repeated desaturations, she required three days of ICU level care before transfer to the general floor. Additionally, she also required 2L of O₂ via oxygen tent at night to maintain oxygen saturation. Twice daily distraction began on postoperative day 5. Apneic episodes leading to hypoxia continued to require stimulation for resolution until day 7. At this time, she continued to have nightly desaturations but began recovering oxygenation saturation spontaneously. She was discharged with a home oxygen tent and continuous pulse-oximetry on postoperative day 11. At her first follow-up appointment on postoperative day 15, she was no longer requiring supplemental oxygen. Figures 1A and 1B demonstrate pre- and post-distraction midface positioning respectively with significant improvement in midface positing after completion of distraction.

OPEN IN VIEWER

Figure 1.

(A) Lateral radiograph of patient 1 prior to distraction initiation. (B) Lateral radiograph of patient 1 during post-distraction consolidation phase. (C) Clinical photograph of patient 1 prior to procedure. (D) Postoperative photograph of patient 1 with halo distractor in place after completion of distraction phase.

Three months after discharge, she was reevaluated by polysomnography and her postoperative AHI was found to be 5.8 indicating a reduction from severe to mild OSA.

Discussion

Postoperative respiratory status appears to be highly heterogenous amongst the syndromic craniofacial dysostosis patient population. While much is known on the proper management of OSA in patients with syndromic craniosynostosis, there is a paucity in the literature regarding the management of respiratory compromise in the immediate postoperative period following LeFort III with external distraction hardware placement.

It cannot be definitively predicted which patients will experience significant postoperative respiratory compromise. However, based upon postulation from this patient cohort, several patient factors should prompt clinicians to preemptively prepare for respiratory intervention following surgery. One such feature would be the preoperative diagnosis of Pfeiffer syndrome, as severe life-threatening OSA is known to occur more frequently in this population. Additionally, the severity of OSA based upon AHI does not seem to be a predictor of postoperative respiratory compromise, as two patients in this cohort (Patients 2 and 4) with preoperative AHI >20 experienced no significant complications.

Several noninvasive management strategies may be employed to manage nighttime desaturation in patients with continued OSA during the immediate interval following LeFort III but prior to initiation of distraction. While the majority of these patients did well in the postoperative period on either room air or via oxygen tent, the one patient who experienced significant desaturation required extremely close monitoring and frequent stimulation. Patient 1 additionally had a prolonged hospital course and ultimately required discharge with supplemental oxygen, though nightly desaturations did slowly improve with initiation of inpatient distraction. Limitations of this study include small patient size within the data collection period and lack of generalizability.

Conclusion

The high incidence of OSA in syndromic craniosynostosis is both well-known and described. Though multifactorial, one major contributing factor is midface hypoplasia. ¹² Correction of this midface hypoplasia can increase both the length and width of the nasopharynx leading to significant improvement and often resolution of OSA.^13,14 In a 2018 meta-analysis of 24 papers on LeFort III advancement in syndromic craniosynostoses, there was no mention of management of perioperative respiratory compromise due to obstructive episodes prior to the distraction period. ¹⁵ Despite many patients experiencing acute improvement of OSA postoperatively, the present work demonstrates the marked heterogeneity of acute postoperative respiratory status. Therefore, it is imperative that patient expectations be managed throughout the perioperative period. Given the lack of literature detailing appropriate evidence-based management of these patients, the authors recommend close contact with PICU care team regarding challenges of airway management including training in the partial removal of hardware to allow for rapid intubation in the event of any severe respiratory event. Additional recommendations include utilization of a face tent and prolonged ICU care. Finally, sleep medicine should be consulted for home oxygen and monitoring. The present work highlights the insufficiency of current recommendations for management of more severe respiratory compromise in the period immediately following LeFort III for the treatment of OSA secondary to syndromic craniosynostosis. Future studies are needed to 1.) further investigate predictive factors for exacerbated sleep-disordered breathing postoperatively, 2.) examine the utility of the oxygen tent and its effect on length of ICU level and total hospital length of stay, and 3.) determine an evidence-based protocol for managing acute postoperative respiratory compromise following LeFort III for syndromic craniosynostosis.

Supplemental Material