Scientific Abstract Presentations: Craniomaxillofacial 1 Thursday, October 26 Thu, Oct 26 1:30-2:30 p.m. CDT

PURPOSE: A review of the Global Burden of Disease study reports a 43% global increase in reported facial fractures in 27 years. Unfortunately, there is limited research investigating population's demographics and facial fracture's characterization. The purpose of this study is to conduct one of the first national evaluations of the prevalence and demographics of co-occurring facial fractures. METHODS: This was a database study of deidentified, aggregate data using PearldiverTM. Between 2010 and 2020. The database was queried by CPT codes categorizing by fracture patterns and surgical treatment (open reduction internal fixation (ORIF) vs. closed reduction). Concurrent treatment of two or more fracture types was determined based on CPT codes reported concomitantly within 30 days of each other. Analysis was conducted utilizing PRISM software. Venn diagrams were used to visually compare the percentage of patients with two-fracture pattern respective to their individual singular-fracture groups. RESULTS: One-fourth of the 244,751 patients identified were between ages 15-24 years old. About 75% of patients had commercial insurance. Closed reduction of nasal fractures was the most common surgical intervention (56%). 22% of malar ORIF patients had a second orbital fracture, but only 2.5% of all orbital fracture patients had a second malar fracture. 35% of NOE ORIFs have a malar fracture, and 2.5% of malar ORIFs have a concomitant NOE fracture. Our results contradict conclusions by Buchanan et al. from the Seattle Program published in 2012.¹ They described 24.5% of malar fracture patients likely having a concomitant NOE fracture, which is nearly ten times more prevalent than our data reports. About 1/3 of either Le Fort I or II ORIFs had a concurrent malar ORIF. Contrasting, 3-8% of malar ORIFs had a concomitant Le Fort I or II fracture. About 1/3 of all Le Fort fractures had a concomitant orbital floor blowout fracture (BOF). Contrasting, only 5% of patients with an orbital floor BOF had any Le Fort fracture. Nasal and mandibular ORIFs had isolated fracture patterns. CONCLUSION: 30% of any Le Fort, malar, or orbital fracture needing ORIF treatment will have a second concomitant facial fracture needing surgical reduction, suggesting surgeons to have a higher index of suspicion for a secondary facial fracture in patients initially identified with any Le Fort, malar, or orbital fracture. 35% of NOE fractures have a malar fracture and 2.5% of malar fractures have NOE fracture. This directly contradicts current literature. Surgeons may consider looking for a second malar fracture in patients identified with an NOE fracture, but not vice versa. Majority of nasal and mandibular fractures occur in isolation. Surgeons may not need to look for a secondary facial fracture in patients with an initial identified mandibular or nasal fracture. REFERENCES: ¹Buchanan EP, Hopper RA, Suver DW, Hayes AG, Gruss JS, Birgfeld CB. Zygomaticomaxillary complex fractures and their association with naso-orbito-ethmoid fractures: a 5-year review. Plastic and reconstructive surgery. 2012;130(6):1296-1304.

Purpose: Secondary alveolar bone grafting (SABG) during mixed dentition is the standard of care for patients with complete cleft lip and palate. Early SABG (4-7 years) occurs before the eruption of lateral incisors, while late SABG (8-12 years) occurs before the eruption of maxillary permanent canines. This study compares outcomes of early ABG (E-ABG) versus late ABG (L-ABG) among patients with unilateral cleft lip and palate (UCLP).

Methods: A retrospective review was conducted evaluating non-syndromic patients with UCLP who underwent ABG from April 2018 to January 2020. Patients with preoperative and postoperative cone beam computed tomography (CBCT) imaging of at least six months from the index operation were included. Demographics, age of surgery, perioperative data, and periodontal information were collected. Preoperative cleft width, bony bridge formation and thickness, incisor root length, and periodontal bone height on cleft-incisor were assessed by Dolphin Imagine software. The Bergland scale score using three-dimensional CBCT rather than traditional occlusal radiograph was applied to assess bone graft outcomes.

Results: After 340 patients with UCLP were screened, 49 patients were included, of which 21 were in the E-ABG group (6.8±1.0 years of age) and 28 were in the L-ABG group (10.8±1.6 years of age). The initial alveolar cleft width is significantly smaller in E-ABG group compared to the L-ABG group (5.4±1.9mm vs. 6.6±2.0mm; p=0.035). However, L-ABG group had higher graft failure rates (32.1%) compared to the E-ABG group (32.1% vs. 14.3%; p=0.150). The overall Bergland scale scores were 1.7±1.0 and 2.5±1.2 for the E-ABG and L-ABG cohorts, respectively (p=0.009). Compared to the E-ABG group, the L-ABG group had significantly greater bony bridge thickness (6.0±2.2mm vs. 3.5±1.8mm; p<0.0001), longer post-graft incisor root length (8.9±2.6mm vs. 13.3±1.8mm; p<0.001), and greater periodontal bone coverage on the root of the cleft-adjacent incisor (80.9±18.6% vs. 66.7±19.9%; p=0.029).

Conclusion: Our findings suggest that patients who undergo early SABG at 7 years may have better graft outcomes and benefits to the periodontal bone support on cleft-adjacent incisor compared to late SABG at 11 years. Existing literature also support early SABG due to favorable outcomes compared to late ABG.(1, 2) Further investigation is critical to determine the optimal timing of ABG.

References: 1) Fahradyan A, Tsuha M, Wolfswinkel EM, Mitchell KS, Hammoudeh JA, Magee W 3rd. Optimal Timing of Secondary Alveolar Bone Grafting: A Literature Review. J Oral Maxillofac Surg. 2019 Apr;77(4):843-849. doi: 10.1016/j.joms.2018.11.019. Epub 2018 Nov 26. PMID: 30576671. 2) Doucet JC, Russell KA, Daskalogiannakis J, Mercado AM, Hathaway RR, Semb G, Shaw WC, Long RE Jr. Early Secondary Alveolar Bone Grafting and Facial Growth of Patients with Complete Unilateral Cleft Lip and Palate. Cleft Palate Craniofac J. 2022 Feb 16:10556656221080990. doi: 10.1177/10556656221080990. Epub ahead of print. PMID: 35171057.

Purpose The SARS-CoV-2 (COVID-19) pandemic was a major disruptor of healthcare, with aftershocks still being felt today. Importantly, pediatric diagnoses of COVID represented up to 20% of all reported cases since the beginning of the pandemic. Given its tropism, it is important to understand if COVID-19 affects outcomes in procedures involving the respiratory tract. The purpose of this study was to determine the risk of respiratory complications in children undergoing cleft palate repair.

Methods This is a retrospective study conducted with the TriNetX database. This database aggregates cases from 55 international healthcare organizations. Pediatric (age ≤3 years) patients undergoing palatoplasty cases based on CPT codes were included if they had a history of COVID-19 (HxC19) based on ICD codes Z86.16 or U07.1. Patient demographics, medical comorbidities, and complications were compared using single variable analysis. Cases were then propensity score-matched within the database to assess for risk of respiratory complications. All statistical significance was set to p < 0.05.

Results There were 1,114 cases of pediatric palatoplasty meeting inclusion and exclusion criteria: 1,054 cases were without HxC19, and 90 cases reported a diagnosis of COVID-19. Prior to propensity matching, cases without HxC19 had a higher risk of acute respiratory distress (risk ratio [RR], 6.16; P<0.0001), stridor or wheezing (RR, 4.33; P<0.0001), hypoxemia (RR, 11.7; P<0.0001), and emergency reintubation or initiation of ventilation assist (RR, 11.7; P<0.0001). Propensity matching of demographics, comorbidities, and cleft diagnosis, yielded cohorts of 87 cases each. There was no significant difference between these two cohorts in regard to any of the respiratory complications included for analysis.

Conclusions Given the similar rates of acute airway and respiratory complications in pediatric palatoplasty cases both with and without HxC19, and no significant change in risk following matching, palatoplasty in children with HxC19 remains a safe option. However, patients and parents with a history of COVID-19 should still be counseled on the possible airway complications but can be reassured that a prior diagnosis of COVID-19 is unlikely to affect their outcomes.

Purpose: Mandibular distraction osteogenesis (MDO) is rapidly becoming a standard of care for management of patients with severe Pierre Robin Sequence. The tongue is brought forward to alleviate the airway obstruction. This study will look at an institutional, multi-surgeon experience with MDO over ten years. Methods: This study was conducted as a retrospective chart review including all patients who underwent MDO at the authors' institution from 2012 to 2022. Demographics, preoperative and postoperative respiratory and feeding status, and distraction data were collected. Primary outcomes were achievement of full oral feeds, avoidance of a gastrostomy tube (GT), avoidance of a tracheostomy, discharge from hospital on room air and complications. A significance value of 0.05 was utilized. Results: Twenty-eight patients met inclusion criteria. Three craniofacial surgeons performed all interventions. The average age at MDO was 135 days, mean activation phase was 13.6 days, mean distraction length was 14.9 mm and mean consolidation phase was 64.2 days. MDO showed a significant improvement in apnea-hypopnea index (AHI) (p < 0.001) and %O2 Nadir (p < 0.001) on postoperative polysomnography. Syndromic patients were associated with discharge with a GT and negatively associated with avoidance of tracheostomy. Patients with preoperatively diagnosed tracheomalacia and subglottic stenosis were negatively associated with avoidance of a tracheostomy. Preoperative diagnosis of gastroesophageal reflux disease (GERD) was negatively associated with extubation immediately to room air (RA) following index procedure. A shorter duration of intubation following index procedure was associated with discharge on RA and a longer duration of intubation was associated with unilateral pan-facial nerve palsy. A longer activation phase was associated with discharge with a GT and a shorter activation phase was associated with discharge on full oral feeds. The ability to discharge on RA was associated with a shorter latency phase, shorter activation phase, and decreased distance of distraction. Repeat MDO was associated with postoperative infection requiring operative management. Conclusions: The goal of MDO is to achieve full oral feeds with no respiratory support. Several interesting findings are described in this study. When assessing patients preoperatively it is important to note the higher rate of failure to avoid a tracheostomy in patients with a syndrome and those with tracheomalacia and subglottic stenosis noted on airway evaluation. Syndromic patients were also noted to be more likely to be discharged with a GT. Several different latency periods were used in this study, and it appears that not only is a short latency period safe, but that it may be correlated with successful discharge on RA. This should be tempered with the finding that a decreased time spent intubated following surgery, decreased distance distracted and time spent in activation was also associated with discharge on RA, which may be a function of the severity of micrognathia in these patients. Of note, patients that spent a longer period intubated following surgery experienced a higher rate of facial nerve palsy. The reason for this correlation is unclear but may represent more severe cases that resulted in more forced retraction during surgery.

Introduction: Infantile hemangiomas (IH) are benign vascular tumors in infants, that rapidly grow in the first year of life and involute during childhood.¹ Propranolol is the standard of care due to the ability to inhibit proliferation, induce regression, and result in involution of the IH.² Despite the minor side effect of propranolol induced sleep disturbance, sleep disruption constitutes the most common reason for early medication discontinuation by parents.³ This prospective pilot study aimed to compare the impact of propranolol therapy on sleep when the drug is administered in either BID or TID dosing regimens.

Methods: This was a prospective single center pilot study. Patients were assigned to one of three therapy groups: BID propranolol (oral), TID propranolol (oral), or timolol (topical)). Patients with multiple hemangiomas or a single hemangioma >2cm were randomized into either BID or TID dosing. Patients with an isolated hemangioma <2cm were prescribed timolol, a topical beta-blocker, as the control. Parents were offered the Brief Infant Sleep Questionnaire (BISQ) at each clinic visit (every 3 months) until termination of the medication (18 months of age).⁴ In the BISQ, scores ranged from 1-5 with 1 representing no changes in sleep and 5 representing major changes in sleep. A univariate analysis was performed.

Results: A total of 94 BISQ surveys were given to 58 patients: 47 at the time of initiation and 46 at the time of follow up. Patients were assigned to treatment group with 56 in the BID group, 17 in the TID group, and 20 in the timolol group. The BID group had the least nighttime sleep hours with 8.08 ± 2.44 compared to the TID group with 9.05 ± 2.32 and the timolol group with 9.50 ± 1.31 (p=0.23). Subjectively, parents in the BID group reported worse sleep scores at 1.84 ±1.29 compared to the TID group at 1.71 ± 1.1 and the timolol group at 1.38 ± 0.52 (p=0.07). Patients in the BID group took the fewest naps per day at 2.72 ± 1.31 compared to 3.29 ± 1.90 and 3.38 ± 1.70 in the TID and timolol groups, respectively (p=0.27). After taking propranolol for 30 days, patients in the BID group exhibited significantly fewer naps per day (Δ=-0.70 naps, p=0.04) and less total time spent napping (Δ=-1.93 hours, p=0.01).

Discussion: Alternative dosing of propranolol appears to play a role in the sleep patterns of infants. BID dosing had significant decreases in total naptime and naps per day, differing from the normal development of infant sleep patterns. The BID dosing pattern additionally had the worst nighttime sleep, sleep scores, and fewest naps per day. The novel TID dosing strategy exhibits sleep patterns consistent with the control group indicating less overall sleep disturbance. Improvements in sleep patterns are likely to have increases in long term drug adherence by parents leading to maximum treatment efficacy. Ongoing enrollment of patients is crucial to eliminate data skew and provide better recommendations in the dosing regimens for beta blockers in the management of infantile hemangiomas.

1. Léauté-Labrèze C, Harper JI, Hoeger PH. Infantile haemangioma. The Lancet. 2017;390(10089):85-94. doi:10.1016/s0140-6736(16)00645-0.
2. Léauté-Labréze C, Hoeger P, Mazereeuw-Hautier J. A randomized controlled trial of oral propranolol in infantile hemangioma. Journal of Vascular Surgery. 2015;62(2):518-519. doi:10.1016/j.jvs.2015.06.197.
3. Ji Y, Chen S, Wang Q, et al. Intolerable side effects during propranolol therapy for infantile hemangioma: Frequency, risk factors and management. Scientific Reports. 2018;8(1). doi:10.1038/s41598-018-22787-8.
4. Del-Ponte B, Xavier MO, Bassani DG, et al. Validity of the brief infant sleep questionnaire (BISQ) in Brazilian children. Sleep Medicine. 2020;69:65-70. doi:10.1016/j.sleep.2019.12.018.

Purpose: Facial aging is a multifactorial process involving both soft tissues and bony structures1. Factors including volume loss, gravity, muscle laxity, and cellular damage contribute to decreased skin and soft tissue elasticity, resulting in increased mobility of facial soft tissues1-2. Variations in soft tissue integrity have been attributed to age, sun exposure, and smoking. While facial aging has been extensively examined histologically, the present study sought to leverage clinical MRI to quantify facial soft tissue movement (STM) and correlate with environmental and demographic factors.

Materials & Methods: Sixty-eight patients underwent high resolution MRI scans, which included two identical scans at the beginning and end of imaging separated by approximately 45 minutes. MRIcron was used to label 49 reproducible bony and soft tissue facial landmarks on all scans. An avatar scan was used to co-register and scale all scans. For each patient, early and late scans were coregistered, and mathematical voxel-wise absolute differences were used to render composite maps to highlight the change in soft tissue configuration over the 45 minute gap. Lines between close neighbor landmarks offered corresponding paths by which movement could be compared between patients. Linear regression was used to correlate average absolute differences with age, sex, smoking status, and sun exposure.

Results: Age was positively correlated with the greatest STM compared to sun exposure, sex, and smoking status. In the upper face, age was correlated with STM in the forehead (glabella to superior orbits, p=0.026), bony orbits (p-range=0.001-0.023), and orbital soft tissue (orbits to medial/lateral canthi, p-range=0.001-0.014). Age was associated with bilateral midface STM in the infraorbital region (between malar eminence, bilateral canthi and inferior orbit, p-range=0.001-0.019) and zygomatic region (malar eminence to auditory canal, p-range=0.001-0.032). In the lower face, age correlated with STM around the mouth and philtrum (lips, oral commissures, columella and nares, p-range=0.028-0.001), and between the mandible and mentum (p-range=0.001-0.031). Sun exposure was only associated with STM in the oral region (lips, columella, and nares, p-range=0.001-0.049) and infraorbital/nasal region (nares to medial canthus, p=0.001). Sex was only associated with STM around the mandibular angle (p=0.004). Smoking was not found to be associated with significant bilateral STM.

Conclusions: To our knowledge this is the first study to examine facial STM using clinical in vivo MRI. This methodology identified facial regions most susceptible to changes from aging and various environmental factors. These results provide further understanding of the natural facial aging process and may be helpful in identifying rejuvenative targets in the future.

References: 1. Ilankovan V. Anatomy of ageing face. Br J Oral Maxillofac Surg. 2014 Mar;52(3):195-202. doi: 10.1016/j.bjoms.2013.11.013. Epub 2013 Dec 23. PMID: 24370442. 2. Freytag L, Alfertshofer MG, Frank K, et al. Understanding Facial Aging Through Facial Biomechanics: A Clinically Applicable Guide for Improved Outcomes. Facial Plast Surg Clin North Am. 2022 May;30(2):125-133. doi: 10.1016/j.fsc.2022.01.001. PMID: 35501049.

PURPOSE Craniofacial reconstruction or enhancement requires autologous, cadaveric or alloplastic implants, often requiring intraoperative modification to optimize shape and fit for a particular patient, introducing associated risks of infection, extrusion, expense and donor site morbidity. Given these shortcomings, there is substantial need for biocompatible, customizable and low-cost facial implants. This study examines the biocompatibility and durability of various polylactic acid (PLA) implant designs +/- decellularized cartilage infill to assess translatability to craniofacial (nasal) reconstruction.

METHODS & MATERIALS Cartilage harvested from ovine ribs was minced or zested and decellularized. Dorsal nasal scaffolds with contours similar to commercially available silicone dorsal nasal implants were custom-designed with 3D modeling software and printed in polylactic acid (PLA) on a PRUSA i3 MK3S+ printer. All scaffolds were heterotopically implanted on rat dorsa with 4 implants per rat. Two scaffolds designed as external "cages" of low or high porosity and were without internal supports. These cages were implanted empty or with processed decellularized cartilage (either "minced" or "zested") infill, yielding 6 total treatment groups, n=4 each. The remaining two scaffolds were designed with internal PLA rebar supports and implanted without decellularized cartilage infill. Scaffolds were explanted 3, 6 and 12 months after implantation, after which they underwent same-day volumetric analysis via microCT. Explants were then Formalin-fixed, embedded in paraffin and sectioned for histopathologic and immunohistochemistry analysis. Statistical and image analysis was completed on RStudio and ImageJ.

RESULTS Overall, implant contours were best retained across all timepoints in the low and high porosity rebar groups. Empty and zested cartilage-filled cages had significant volume loss at 6 and 12 months relative to volume at implant (p < 0.05) with explanted constructs grossly collapsed at both timepoints. Rebar and minced cartilage-filled constructs had superior volume retention relative to empty and zested cartilage-filled cages at 6 months; only rebar groups retained this volume after one year (p < 0.05). Uniform and well-distributed neotissue ingrowth was evident in all filled implants on MicroCT; "cage" implants without infill no longer had a measurable internal cavity volume due to flattening. Hematoxylin and eosin and safranin-O stain showed a robust lymphocytic inflammatory response at 3 months that subsided by 6 and 12 months, as well as increased tissue vascularization and collagen deposition between timepoints consistent with healthy tissue ingrowth. Immunostaining indicated an environment favorable to tissue proliferation with more M2 than M1 macrophages across all timepoints and strong M2 expression among the cells bordering PLA.

CONCLUSION This study supports the use of PLA for generating customizable facial implants. With sufficient internal supports, constructs retain contours and volume at one-year post-implant and provide a scaffold for ingrowth of healthy, vascularized, collagen-rich tissue. While volume retention for decellularized cartilage-filled cages was inferior to PLA-only groups, these findings support the biocompatibility of combined decellularized cartilage-PLA implants and future designs may incorporate both rebar and cartilage infill. Importantly, design and fabrication of these bespoke implants can be completed quickly and at minimal cost, potentially at the point of care, allowing for affordable implants with a minimal risk profile.

Background: Head and neck lymphedema (HNL) after head and neck cancer treatment can significantly impact patient quality of life by causing difficulty with swallowing and speech, increasing cost of treatment, leading to facial disfiguration, and decreasing psychosocial wellness. Currently, The M. D. Anderson Cancer Center (MDACC) HNL evaluation protocol is the most commonly used to tool to evaluate the severity and status of head and neck lymphedema by utilizing a series of face and neck tape measurements. However, there is no gold standard assessment tool that reliably and consistently measures HNL despite over 35 assessment tools having been reported in the literature to date. With the continued integration of advanced technology in medicine, three-dimensional (3D) imaging has shown that it's an effective and reliable tool to assess volume changes. Therefore, we studied the efficacy of 3D imaging as a novel method of tracking changes in head and neck volume in patients with head and neck lymphedema. The aim of this study was to compare changes in 3D measurements against traditional MDACC protocol tape measurements.

Methods: Patients undergoing treatment for head and neck cancer were prospectively enrolled, and 3dMD images were captured at the initial and each subsequent appointment. The MDACC protocol was also used to obtain a composite bilateral hemifacial score from the sum of all facial tape measurements at each appointment. Additionally, the MDACC rating scale was used to determine the degree of lymphedema. Following study completion, the changes in 3dMD imaging measurements and bilateral hemifacial scores from first to last visit were compared for each patient. Pearson's correlation coefficients were used to assess the strength and direction of the relationship between the two measures.

Results: A total of 230 patients were included in the study. The mean age at time of first visit was 61.5 ± 9.4 (SD) years. Most primary cancers originated from the oropharynx (35%), oral cavity (26%), or larynx (17%), with the remaining 22% originating in other locations. 31% of patients received radiotherapy, 8% received surgery, and 61% received both radiotherapy and surgery. 20% of patients were categorized as level 1a lymphedema, 71% as 1b, and 9% as level 2. The median time from first visit to last visit was 3.3 months (2.1-5.8, interquartile range). The average changes in lymphedema measurements were 1.9 ± 25.8 using 3dMD scans and -1 ± 5.2 using the MDACC protocol. The Pearson correlation coefficient for these two variables was 0.207 (p = .002), indicating that there is a low, positive correlation between tape and 3D measurements.

Conclusions: Use of 3D imaging for volume measurement in head and neck lymphedema has not been previously studied. Our study demonstrates that while traditional tape measurement is frequently used to measure HNL, efficacy and accuracy are low. The reliability and accuracy of 3D volume measurement has been validated in other studies, and we believe it is an effective tool in evaluating head and neck lymphedema.

Purpose: Movement of the bony orbits can be accomplished with box osteotomy (BO) or facial bipartition (FB). Both procedures have been shown to successfully reduce the interdacryon distance (IDD) as well as the overlying soft tissue, but little data exists to support the use of one procedure over the other. This study compares the outcomes of BO and FB at a single center and proposes an algorithm to assist in preoperative decision-making.

Materials and Methods: A retrospective review of patients undergoing BO or FB was performed at a single institution from 2005 to 2022. Patient demographics, medical history, perioperative data, length of hospital stay (LOS), and postoperative complications were collected and analyzed. Correction of the IDD in BO compared to FB was measured on pre- and post-operative CT scans. Chi-squared and Mann-Whitney U tests were used for statistical analysis.

Results: Forty-one patients were included, 27 FB and 14 BO, with varying diagnoses. In the patients with hypertelorism, there was significant improvement of the IDD after surgery for both BO patients (preoperative: 34.8±6.3 mm vs. postoperative: 25.7±6.3 mm; p=0.018) and FB patients (preoperative: 27.9±5.6 mm vs. postoperative: 21.6±2.8 mm; p<0.001). Notably, the BO cohort had a significantly larger reduction in the IDD compared to the FB cohort (10.2±1.7 mm vs. 6.4±4.4 mm; p=0.014). One (7.1%) BO patient had a complication whereas 13 FB patients (18.5%) had complications (p=0.009). One FB patient required reoperation for orbital dystopia. The BO cohort had significantly lower intraoperative blood loss (400 vs. 825 mL; p=0.007), lower transfusion requirements (293 vs. 799 mL; p<0.001), and shorter LOS (5.5 vs. 8.0 days; p=0.005) compared to the FB cohort. Average follow-up time was 7.1±4.7 years (FB) and 3.0±2.6 years (BO; p=0.005).

Conclusion: BO and FB are effective operations for correcting orbital malposition. Both procedures are safe when performed by an experienced surgeon in a tertiary care facility. A surgeon must take into consideration other surgical needs to correct the facial cascade such as malar prominence position and palatal arch width when determining which procedure to perform.