The role of Nasoalveolar molding: A 3D Prospective analysis

Nasoalveolar molding (NAM) is commonly employed to reduce the alveolar segments into proper alignment and to improve nasal symmetry in patients with cleft lip and palate. This study examines the periodical progression of NAM treatment over time. 20 patients with complete unilateral cleft lip and palate were prospectively recruited. A 2 stage NAM treatment protocol was applied. Stage 1 involved adjustment of the alveolar segments (mean age 15.6 days), while Stage 2 added nasal stents and started average 43 days after stage 1. 3D images (n = 241) were obtained prior to NAM initiation and weekly until the end of treatment. The cleft lip area, bilateral nostril areas, and the nostril height and width were measured. Treatment was assessed in the Cleft (C) side and the Non-cleft (N). There was significant difference in the C/N ratio of the nostril area, width, and height at pre-treatment (0.9 ± 0.3, 4.1 ± 1.1, and 0.5 ± 0.2), at the end of stage 1 (1.1 ± 0.3, 2.2 ± 0.6, and 0.8 ± 0.2), and at the end of stage 2 treatment (1.8 ± 0.3, 1.8 ± 0.4, and 1.2 ± 0.1); p < 0.05. Comparative 3D analysis with dense sampling offers a precise methodology for showing effects of NAM treatment. The morphological changes achieved with NAM therapy occur in early treatment phase.

Overall comparisons. On average, the cleft nostril area increased in direct proportion to duration of NAM treatment. During the first stage of NAM, the cleft nostril area increased at the same rate as the non-cleft side nostril but during the second stage it increased significantly faster than the non-cleft side. The non-cleft side nostril area increased only during the 1 st NAM stage. The cleft nostril height increased over the entire duration of NAM therapy, while the non-cleft nostril height increased predominantly in stage 1. The major decrease of the cleft nostril width occurred in stage 1, while the non-cleft nostril width increased slowly over the full span of treatment (Fig. 2).
The differences in outcomes by stage of treatment were tested for significance by calculating ratios of cleft to non-cleft values (C/N) for nostril area, height, and width (Fig. 3). These ratios were found to be statistically different between each stage of treatment for all outcome measures. The C/N ratio of nostril area and nostril height improved more in the second stage of therapy, while nostril width increased more in the first stage than in the second.
Group comparisons. There were 9 patients in group 1, 8 patients in group 2 and 3 patients in group 3. The C/N ratios of nostril area at post stage 2 are 1.8 ± 0.3, 1.7 ± 0.3, and 1.7 ± 0.2 in group 1, 2, and 3. The C/N ratios of nostril height at post stage 2 are 1.2 ± 0.1, 1.2 ± 0.2, and 1.2 ± 0.1 in group 1, 2, and 3. The C/N ratios of nostril width at post stage 2 are 2.0 ± 0.3, 1.8 ± 0.4, and 1.6 ± 0.2 in group 1, 2, and 3 (Fig. 4). Owing to the little case number in group 3, no statistics are performed for the comparisons of the three groups in different stages. Even though, the C/N ratios of the three groups are much different in the pre-NAM treatment and post-stage 1, the ratios are similar at the end of the NAM treatment, indicating equivalent outcome.

Discussion
3D imaging techniques have been widely adopted by the craniofacial community and have been shown to be accurate when compared to manual anthropometry 16,22,23 . However, relatively fewer 3D studies related to NAM have been published 6,15,24 and 2D methods are more numerous [9][10][11][12][13]25 . To our knowledge, this is the first study to prospectively and rigorously analyze the 3-dimensional progressive changes with dense acquisition of periodic imaging data in children with unilateral complete cleft lip and palate who underwent presurgical NAM treatment prior to primary lip repair.
The overarching goal of this study was to understand how rapidly NAM produces changes. During the first stage of treatment, nearly 50% reduction in the diastasis of the lip was observed in the first two weeks. Stability was reached at 6 to 7 weeks of treatment when the second NAM appliance was delivered, as the treatment focus shifts to molding nasal cartilage and lengthening of the columella. The rapid decrease in the cleft nostril width occurs concomitantly with approximation of the alveolar segments during the first stage of NAM treatment. With alveolar segments alignment, the distance between alar bases and the diastasis of the lip are reduced. Yet it was intriguing to observe that the nostril width asymmetry between cleft and non-cleft sides continued to decrease in the second stage of NAM treatment. It seems that this ongoing decrease in cleft nostril width during the second phase results from reshaping of the nasal cartilages towards normal. Similarly, there was significant correction of nostril height during the first stage, although correction of nostril height is considered a primary treatment goal of the second stage, when nasal stents are employed as a feature of the appliance. These results indicate that alveolar segment molding not only reduces the distance between segments, but also elevate the immature and flat lower lateral cartilage on the cleft side.
Significant changes in nostril area occurred in the first and second stages of NAM treatment. In the first stage, non-cleft side nostril area increases as the alveolar segments are approximated and lip diastasis decreases. In addition, the cleft nostril area increases as the cleft nostril height increases despite the decrease in cleft width which occurs in this phase. In the second stage, non-cleft nostril area remains stable but cleft side nostril area increases. This occurs as the cleft nostril height increases while the width remains nearly unchanged.
Noting that significant changes occur at early stage of NAM treatment prompts the question of the advantage of initiating the NAM treatment at young age. In this preliminary study, we noticed that children older than a month of age at initiation of NAM had more asymmetry with respect to nostril area after stage one, which could be attributed to the relatively shorter stage one treatment for this age group. Overall, there were no significant differences in cleft to non-cleft ratios at completion of therapy. Matsuo reported that delay in NAM treatment might influence outcome due to decreasing plasticity in the nasal cartilage as the infants age 26 . Shetty concluded most of the benefit of NAM is obtained when NAM is applied before 1 month of age 12 . In our study the three patients over a month in age at time of commencing therapy eventually obtained equally good outcome as the other children who had earlier initiation. Again, further study is warranted due to small sample size of this age grouping. Despite the good outcome in these patients, we concur that earlier treatment is prudent 27 . We believe that earlier treatment of congenitally deformed cartilage structures favors earlier molding intervention, as has been established in neonatal ear cartilage molding 28,29 .
For nasoalveolar molding therapy to succeed (and for complete data collection in this study), parent and caretaker compliance with frequent visits is crucial 30,31 . Chen et al. suggested that reduced frequency of adjustments results in decreased cost of care 11 . In our study, we did not test adjustment frequency but the methodology used here would be applicable in a well-designed prospective study. In contrast to our progressive analysis with dense sampling, most of the literature compares only the pre-and post-NAM morphologies 9, 10, 12, 27 . Our study demonstrates the tempo and degree of progressive changes in nasal symmetry, and was able to document complete nasal correction or even overcorrection prior to primary lip repair. According to van der Heijden et al., the evidence that NAM achieves nasal symmetry is not compelling 32 . On the other hand, Liou et al. suggests overcorrection to compensate relapse and differential growth after primary lip repair 33 . While this study does not address the postsurgical value of preoperative NAM, the generally agreed upon presurgical goals of NAM treatment (i.e. reduction in the size of alveolar gap and improvement in nasal symmetry) were achieved in this study 34 . Long-term follow up of the cleft patients receiving the NAM treatment prior to lip repair should be observed to ensure the durability of the improvements noted with NAM treatment. Further studies might be focused on enrolling a control group to tease apart the effects of growth vs. NAM treatment, and assurance that adequate statistical power is achieved.
Serial 3D analysis based on dense samplings provides a precise methodology for evaluation of NAM treatment. Early achievement on the morphologic change with NAM therapy could be expected. NAM achieved significant improvement in measures of cleft width and nasal form in all patients, regardless of the initial age of treatment.

Materials and Methods
Institutional Review Board (IRB) approval was obtained by UT Southwestern medical center and all methods were performed in accordance with the relevant guidelines and regulations. In addition, informed consent was obtained for both study participation, and publication of identifying images. 20 patients with unilateral complete cleft lip and palate without other facial deformity were recruited to the study. The children's ages ranged from 1 to 8 weeks old.

Presurgical Nasoalveolar Molding.
A two-stage modified NAM protocol was performed. The patients visited the Craniofacial Team Care Clinic weekly for NAM adjustments.
Preparations for NAM appliance was made during the first appointment, which consisted of (1) parental intake interview, (2) intraoral and extraoral 3D stereophotogrammetric imaging, and (3) intraoral impression with putty (PVS, polyvinylsiloxane) impression material. These impressions were made by the orthodontist in the outpatient clinic without the use of anesthesia. The NAM appliance was applied at the second appointment (7 days post impression). Parents were instructed to leave the NAM appliance in place at all times except for cleaning.
The NAM procedure in our study consisted of two stages. Stage 1 starts with alveolar molding alone, using a first appliance without a nasal stent and with dental floss threaded to the appliance for safety to prevent aspiration.  The appliances were adjusted weekly to mold the alveolar segments by selective grinding and application of soft acrylic. Stage 1 ends when the gap between the cleft alveolar segments became smaller than 5 mm (~6 weeks) or lip segments could be approximated without much tension. Stage 2 began with the use of a second appliance, providing nasal stents to correct the nose projection and asymmetry. The appliance was retained with denture adhesive and the lip is taped with 3 M micropore tape at all times. A small amount of Mastisol was applied to the lip to increase tape retention.
The patients underwent 3D stereophotogrammetric imaging (3dMD, Georgia, USA) weekly (without NAM appliances in place) until the end of treatment (Fig. 5). A total of 241 images were obtained for all patients. Every effort was made to acquire the images with a neutral facial expression. Treatment History. Each patient's clinical notes were examined and details of nasoalveolar molding treatment were recorded, including age at beginning of treatment, presence of complications (e.g., facial irritation or oral mucosal ulceration), and treatment duration.
3D image analysis. The images were converted to stereolithography file format (.stl) using 3dMDVultus software (3dMD, Georgia, USA) to prepare them for analysis. Cleft lip area, nostril area, nostril height and nostril width were measured using 3-Matic software (Materialise, Belgium).
To calculate the area of the cleft diastasis, a contour that surrounds the lip defect region was determined by recognizable landmarks, including: labiale superius (ls), crista philtri landmark (cph), subnasale' (sn') and subalare (sbal) 35 . The gap was filled with a mesh surface and its area was calculated. Nostril area, width, and height were also calculated. The method utilized involved creation of a reference line defined by spanning the alar bases, which also intersected the base of the columella as closely as possible. The cleft nostril area was then measured by closing the cleft nostril above the reference line as it intersected with the arc of the cleft side ala. The non-cleft nostril area was measured by simply inscribing the closed oval defined by its border. Nostril width was determined as the maximal width of the nostril at a parallel line segment above the reference line, while nostril height was measured by the length of a vector from the highest point of the nostril dome and the reference line (Fig. 6). These measurements were obtained weekly throughout NAM therapy.
Patients were divided into three groups according to the age at initiation of NAM treatment: Group 1 was less than 2 weeks of age, Group 2 was 2 to 4 weeks of age, and Group 3 was older than 4 weeks. NAM treatment progression was assessed at three critical intervals: pre-NAM treatment, post stage 1, and post stage 2.
Statistical Analysis. Paired t-tests were performed to compare NAM progression at different stages and treatment age initiation. Reliability testing was performed on 5 patient datasets (50 images). A single observer obtained measurements on separate days with a three-month interval between landmarking sessions. Intraclass correlation coefficient (ICC) indicated a high reliability ICC = 0.9991 with 95% CI from 0.9989 to 0.9992 and a p-value < 0.0001.
Kolmogorov-Smirnov analysis was performed to test our data for normality. The results indicated that we may assume the data are normal (either pre, post 1 or post 2). Statistical analyses were performed using IBM SPSS for Windows, Version 19.0 (IBM Corp., Armonk, N.Y.). The significance level (α) was set at 0.05 (5%).