Predictors of Velopharyngeal Insufficiency After Surgical Advancement of the Maxilla among Young Adults with Cleft Palate

Thesis (Ph.D.)--University of Washington, 2021Introduction: Maxillary hypoplasia refers to undergrowth of the upper jaw and results in malocclusion of varying severity. While the prevalence of severe maxillary hypoplasia in the general population of the United States is 0.3%, the prevalence among patients with a cleft palate with or without cleft lip (CP/L) is more than 40%. The timeline of care for children with CP/L requires frequent follow-up and multiple surgeries throughout their childhood. When maxillary hypoplasia is severe, surgical maxillary advancement at the conclusion of craniofacial growth, typically between 17-21 years of age, is required to achieve functional occlusion. However, anterior movement of the maxilla also changes the relationship between the soft palate and the posterior pharyngeal wall, introducing risk of velopharyngeal insufficiency (VPI), a speech disorder that interferes with speech acceptability and if significant, speech intelligibility. At present, there are two types of surgery commonly done to treat maxillary hypoplasia in children with a cleft palate: distraction osteogenesis (DO) and traditional maxillary advancement (TMA). The results of prior studies that compare the prevalence of VPI in children after undergoing each type of surgery suggest a lower prevalence in patients who underwent DO, but the difference was small and was based on a relatively small number of patients. The goal of this dissertation was: to assess in a larger sample the relation between type of surgery for maxillary hypoplasia and the presence of post-operative VPI; and to investigate different anatomic and speech characteristics in these patients that are associated with risk of developing VPI. Methods: We analyzed data from a cohort study that utilized data from two internal clinical databases as well as the electronic medical record. The first analysis aimed to compare the risk of post-operative VPI between two different surgical approaches to maxillary advancement: distraction osteogenesis and traditional maxillary advancement. The second evaluated type of cleft, magnitude of maxillary advancement, borderline pre-operative VPI, and pre-operative velopharyngeal ratio as predictors of developing VPI after maxillary advancement. Identification of potential confounders was accomplished through a combination of a priori clinical knowledge as well as a data-drive approach. Results: The prevalence of post-operative VPI among those who underwent distraction osteogenesis was 40.5%. After adjusting for type of cleft and magnitude of overjet, this was 47 percent (95% CI for the PR = 0.72-3.05) higher than the corresponding prevalence of VPI in the TMA group. A modest excess of post-operative VPI was also observed in the DO group when those with a unilateral cleft lip and palate were analyzed separately (adjusted PR=1.44, 95% CI 0.56 – 3.67). Neither history of Furlow palatoplasty nor history of any VPI surgery appeared to bear on the size of the association between maxillary surgery type and risk of post-operative VPI. The increased prevalence of VPI among those undergoing distraction osteogenesis was largely confined to patients with a pre-operative VP ratio of less than 0.8. The prevalence of VPI after either type of maxillary advancement surgery was not associated with the magnitude of maxillary deficiency (adjusted PR 1.01 per 1 mm increase in the distance between A-point and the y-axis, 95% CI 0.97 – 1.03). Conditional on syndrome status, those with borderline pre-operative VPI were more likely to present with post-operative VPI than those with normal pre-operative velopharyngeal function (PR=1.57, 95% CI 0.93 – 2.67). There was a suggestion that patients with bilateral and unilateral CLP were more likely to have post-operative VPI compared to those with a cleft palate only (PR 1.59, 95% CI 0.64 – 3.96; PR 1.67, 95% CI 0.66 – 4.27, respectively). Conclusions: Our results suggest modestly poorer velopharyngeal function among patients who underwent DO compared to those who underwent TMA, after accounting for differences in type of cleft and magnitude of overjet. However, our results are not in accord with earlier findings. To date, however, all analyses on this issue are limited by statistical imprecision and additional contributions to the body of evidence are warranted