Cephalometric predictors of hypernasality and nasal air emission

Abstract During times of increasingly recognized importance of interprofessional practices, professionals in Medicine, Dentistry, and Speech Pathology areas cooperate to optimize treatment of velopharyngeal dysfunction (VPD), after primary palatoplasty for correction of cleft palate. Objective Our study aims to compare velar length, velar thickness, and depth of the nasopharynx of patients with unilateral cleft lip and palate (UCLP) with the presence, or absence, of hypernasality and nasal air emission; and to verify if the depth:length ratio, between nasopharynx and velum, would be predictive of consistent hypernasality and nasal air emission (speech signs of VPD). Methodology Cephalometric radiographs and outcome of speech assessment were obtained from 429 individuals, between 6 and 9 years of age, with repaired unilateral cleft lip and palate. Velar length, velar thickness, depth of the nasopharynx, depth:length ratio, scores of hypernasality, and scores of nasal air emission were studied and compared; grouping the radiographs according to presence or absence of hypernasality and nasal air emission. Results For the group with speech signs of velopharyngeal dysfunction (those with consistent hypernasality and nasal air emission), the velums were shorter and thinner; the nasopharynx was deeper and the depth:length ratio was larger than the group without hypernasality and nasal air emission. Velar length was significantly shorter in individuals with consistent hypernasality and nasal air emission (p<0.001) and with history of palatal fistula (p=0.032). Depth of nasopharynx was significantly greater in individuals with consistent hypernasality and nasal air emission (p<0.001). Depthlength ratio was significantly larger in individuals with consistent hypernasality and nasal air emission (p<0.001). A depth:length ratio larger than 0.93 was always associated with speech signs of VPD. Conclusion Estimated with cephalometric radiographs, a depth:length ratio greater than 0.93, between the nasopharyngeal space and the velum, was 100% accurate in predicting hypernasality and nasal air emission after primary repair of unilateral cleft lip and palate.


Introduction
During times of increasingly recognized importance of interprofessional practices, the interdisciplinary care of craniofacial anomalies is essential to improve quality of life and to reduce burden of care for patients and families. Professionals in Medicine, Dentistry, and Speech Pathology areas of healthcare interact to prevent and optimize the treatment of velopharyngeal dysfunction (VPD) and palatal fistula after primary palatoplasty. Not all cleft palate team worldwide, however, have access to imaging assessment of velopharyngeal function, such as videofluoroscopy or nasoendoscopy. 1 Fayyaz, et al. 1 (2019), for example, proposed assessing velopharyngeal competency as part of a "system of classification for defining and describing palatal fistulae". According to the authors 1 , conducting a videofluoroscopic or nasoendoscopic evaluation "would have been a better approach", but it is not always available, leading the authors to propose a clinical judgment regarding velopharyngeal competency, based in the intraoral examination combined to the outcome of speech assessment.
Although some cleft palate teams may lack equipment for videofluoroscopy or nasoendoscopy, Using cephalometric radiographs to study the relationship between the velum and the pharynx is not a novel idea, 5,6,7,8,9,10 but the literature still is not consistent regarding the predictive value of VL, VT, and DN for the management of VPD. Mazaheri, Athanasiou and Long 7 (1994) used cephalometric radiographs to compare VL, VT and DN between groups of patients with different types of CLP, with and without velopharyngeal competence for speech.
The authors studied 85 individuals with cleft lip and palate, at "6 month intervals during the first 2 postnatal years and annually thereafter up to 6 years of age", and reported that the measurements obtained indicated that it would be impossible to predict those individuals who would later require management of VPD. More recently, Silva,et al. 11 (2017) single research site. In addition, the cephalometric radiographs had to be taken before orthopedic and orthodontic management and prior to surgical management of velopharyngeal insufficiency. Routine settings for obtaining cephalometric radiographs to monitor growth, position, and size of skeletal and dental structures were followed during acquisition of the images selected. Exclusion criteria included cephalometric radiographs that were not adequately obtained, for example: images in which the velum was not at a physiological rest, the teeth were not in occlusion, or the head was not positioned adequately.
Images in which the object of interest for our study were not identifiable were also excluded. In children between 6 and 9 years of age (participants of this study), the third molar is not available to be used as a reference, but third molar germs, when present, may be used as reference. In our study, both third molar germs (when present) and the retro molar maxillary tuber area were used as reference to identify PNS. Since the maxillary tuber is usually aligned with the anterior surface of

Results
To evaluate the error of the method, we repeated the measurements 15 days later for 151 (35%) cephalometric radiographs randomly selected. We used the principles defined by Dahlberg (1940) to estimate the method error for each parameter. 16 The differences between the first and the second measurements for VL, VT, DN were not significant with p=0.066, p=0.616, and p=0.806, respectively.
From the 466 radiographs available, we excluded 37 (8%) due to lack of quality of the image. In total, our study included 429 cephalometric radiographs from patients between 6 and 9 years of age (mean  We established overall means and standard  (Table 1).  Considering the widespread availability of still X-ray machines, the use of cephalometric radiographs can enhance both the diagnose of VPD and the study of velopharyngeal morphology due to its non-invasive nature and limited exposure to radiation.

VARIABLES MEAN (SD) mm INTERPRETATION OF FINDINGS
Studies have described the characteristics of structures involved in velopharyngeal function using single lateral still X-rays or cephalometric radiographs. 4,7,8,9,10,21 Subtelny 4 (1957), in particular, reported that a depth:length ratio greater than 0.70 suggested an unfavorable relationship between the velum and the pharynx, indicating that the DN/VL ratio could be used to predict and identify individuals at risk of VPD. Silva, et al. 11 (2017) reported that the sensitivity of the depth:length ratio as an index J Appl Oral Sci. 2021;29:e20210320 6/7 of VPD was 80%. Whereas participants in the study by Silva,et al. 11 (2017)  Considering Subtelny's 4 (1957) normative data for individuals without CLP, we verified that the author's depth:length ratio measures ranged between 0.66 (SD 0.15) at 6 years, 0.70 (SD 0.14) at 7 years, 0.69 (SD 0.14) at 8 years, and 0.66 (SD 0.13) at 9 years.
The mean depth:length ratio of 0.63 (SD 0.17) found in our study for the group without hypernasality and nasal air emission is similar to Subtelny's data, and is predictive of possibility of velopharyngeal closure.
That is, individuals without signs of VPD, in our study, had depth:length ratio measures within a range that corroborates their speech findings (absence of hypernasality and nasal air emission).
Considering the group with speech signs of VPD (N=56), nearly half (46%) presented depth:length ratio interpreted as false negative for presence of VPD (such as DN/VL ration at 0.70 or below). These findings can be partly explained by an enlarged adenoid tissue, which we ignored, and should be addressed in future research. While measures of VL, VT, DN and depth:length ratio can be particularly important in corroborating speech signs of VPD, in CLP Centers without access to the equipment required for nasoendoscopy and videofluoroscopy, the findings of 2D imaging require careful interpretation, considering the high incidence of false positives. Nevertheless, in our study, when the magnitude of the relationship between the nasopharyngeal space and the velum was at a ratio above 0.93, the DN/VL ratio predicted VPD with 100% accuracy.
Finally, one of the advantages of a cephalometric radiograph is its widespread availability, and likelihood of young children compliance. This suggests the possibility of obtaining a still lateral X-ray for speech assessment purposes even earlier than a dental cephalometric radiography. If this is the case, the use of a contrast of barium sulphate, applied to both nares and to the mouth, prior to obtaining the still X-ray/cephalometric radiograph for speech purpose, may help demarcate the structures, improving the identification of the oral and the nasal surfaces of the soft palate. Conclusion A depth:length ratio between nasopharyngeal space and velum greater than 0.93, was 100% accurate in predicting hypernasality and nasal air emission after primary repair of unilateral cleft lip and palate. Velar length was significantly shorter in subjects with consistent hypernasality and nasal air emission (p<0.001) and with history of palatal fistula (p=0.032). Depth of nasopharynx was significantly greater in subjects with consistent hypernasality and nasal air emission (p<0.001). Depth:length ratio was significantly larger in subjects with consistent hypernasality and nasal air emission (p<0.001).