Abstract
Objective
Hard palate osteogenesis after primary palatoplasty is a concept that remains poorly defined. This study aims to investigate the impact of palatoplasty on hard palate osteogenesis.
Design
Retrospective cohort study.
Setting
Tertiary care academic hospital.
Patients
Sixty-six patients who underwent primary palatoplasty followed by postoperative computed tomography (CT) imaging.
Interventions
Primary palatoplasty and postoperative CT imaging obtained at least 1 year after surgery.
Main outcome measures
The primary endpoint is the dimensions of the hard palatal cleft after palatoplasty. Secondary endpoints included fistula formation and the impact of surgical technique and age at palatoplasty on residual cleft size.
Results
No patients achieved complete osseous closure of the hard palate, with an average residual cleft involving 41.4% ± 16.8% of the hard palate. Veau IV patients exhibited the largest residual clefts and highest rates of fistula formation. Larger cleft defects were significantly associated with an increased fistula risk, with a 0.64% increase in fistula formation for every 1% increase in hard palate involvement. Palatoplasty performed before age 1 was associated with significantly smaller residual clefts than surgery after age 2 (P = .03). Patients undergoing Von Langenbeck palatoplasty had smaller residual clefts compared to the 2-flap Bardach technique.
Conclusions
In this study of 66 patients, no patients demonstrated complete osseous regeneration after palatoplasty. Patients undergoing palatoplasty before age 1 and patients undergoing Von Langenbeck palatoplasty had the smallest residual clefts, which may suggest increased osteogenesis at the hard palate in these groups.
Introduction
Primary cleft palate repair is arguably the most important component of orofacial cleft care. The reconstruction of palatal architecture and physiology enables the development of intelligible speech.1-5 Interestingly, aside from gingivoperiosteoplasty, palatoplasty is one of the few components of orofacial cleft care that involves reconstructing a composite soft tissue and bony defect using soft tissue alone. This contrasts with alveolar cleft repair which involves reconstructing a composite soft tissue and bony defect with both soft tissue and bone. The success of alveolar cleft reconstruction is measured by soft tissue repair and bone graft incorporation, whereas the success of cleft palate repair is measured only by soft tissue integrity and muscle physiology. While the effect of palatoplasty on maxillary growth and speech has been extensively studied, the osseous outcomes after palatoplasty remain poorly understood.3-6
Emerging studies have begun to elucidate the biological mechanisms and clinical implications of hard palate osteogenesis following primary palatoplasty. The literature suggests that factors such as mucoperiosteal flap design, surgical closure technique, and patient age play pivotal roles in this process.7-11 Yin et al. reported that 71% of patients undergoing Von Langenbeck palatoplasty exhibited hard palate osteogenesis, predominantly in the region between the premolars and anterior molars. 7 Similarly, Saijo et al. evaluated 29 patients who underwent push-back palatoplasty and demonstrated consistent bony regeneration near the first molar. 8 Choi et al. documented osteogenesis in 30 patients following Furlow palatoplasty, with 2 cases achieving “near-complete” closure of their bony defect. 9 Finally, Prydso et al. investigated 9 children following palato-vomer plasty and observed osseous closure of the hard palate at the fusion site between the palatal and vomerine flaps. 10 While these studies are commendable for their utilization of postoperative computed tomography (CT) imaging to measure hard palate osteogenesis, significant variability in mucoperiosteal flap design, surgical technique, and postoperative evaluation methods limits the author's ability to draw definitive conclusions about the role of palatoplasty in promoting hard palate osteogenesis.
This study aims to investigate the impact of palatoplasty on hard palate osteogenesis. The authors hypothesize that larger osseous defects in the hard palate after palatoplasty are associated with higher rates of fistula formation. Additionally, they hypothesize that surgical technique and age at palatoplasty influence residual cleft dimensions. The primary endpoint is the dimensions of the hard palatal cleft after palatoplasty. Secondary endpoints include fistula formation and the effect of surgical technique and timing of palatoplasty on residual cleft size.
Methods
Following Institutional Review Board (IRB no. 17111) approval, a retrospective chart review was conducted at a single institution, encompassing all patients with a primary diagnosis of cleft lip and/or cleft palate and isolated cleft palate from 1998 to 2022. A waiver of informed consent was granted; as such, no written or verbal informed consent was obtained. Patients were identified using Research Derivative software, a comprehensive clinical repository derived from the electronic medical records of all consenting patients treated at this institution. The database was queried for patients with International Classification of Disease (ICD) codes: 749.1 (Cleft lip), 749.2 (Cleft palate), and 749.00 to 749.04 (Cleft lip with or without cleft palate).
All patients with a primary diagnosis of cleft lip and/or palate and isolated cleft palate who underwent primary palatoplasty at our institution were analyzed. Numerous surgeons performed palatoplasties over the study period, utilizing either the 2-flap Bardach or Von Langenbeck techniques, both of which involve mucoperiosteal flap closure. Additionally, patients with Veau III clefts had unilateral vomer flaps, and patients with Veau IV clefts had bilateral vomer flaps.
Patients were included if postoperative CT imaging of the hard palate was available at least 1 year after surgery. Postoperative CT imaging was obtained incidentally for various clinical indications unrelated to cleft care, including evaluation for sinus disease, facial trauma, altered mental status, hydrocephalus, and otologic pathology. Since routine imaging is not a standard of practice following palatoplasty, this study leveraged existing imaging from patients who underwent CT scans for other craniofacial concerns.
Patients were excluded if they underwent primary palatoplasty at an outside institution, or if postoperative CT imaging was unavailable for review. Patients with a cleft affecting the soft palate only (Veau I) were excluded. Additionally, patients who underwent alveolar bone grafting, gingivoperioplasty, or orthodontic procedures that may have affected the hard palate's growth or position were excluded, as these variables may have independently influenced cleft and palate development.
CT scans were independently analyzed by 2 radiologists (J.H. and S.P.) who were blinded to patient information. In cases where coronal or sagittal images were unavailable, manual reformatting of the CT scans was performed using a multiplanar reformatting/three-dimensional tool within the Picture Archiving and Communication System (PACS) (Sectra Medical, Linköping, Sweden).
The maximum transverse distance of the hard palate and the maximum transverse distance of the cleft defect were measured in the coronal plane (Figure 1). The anatomic landmarks used included the incisive foramen anteriorly, the palatine rugae and alveolar ridge laterally, and the greater and lesser palatine foramina posteriorly. Additionally, the maximum dimension of the residual cleft defect (length or width) was measured in both the sagittal and coronal planes (Figure 2). Supplemental Figures 1A through F show more examples of measurements taken for all included Veau subtypes. The percentage of hard palate involvement was calculated by dividing the cleft's maximum transverse distance by the hard palate's maximum transverse distance and multiplying the result by 100.

Computed tomography (CT). Images showing the measurements of the maximum transverse distance of the hard palate and cleft defect in the coronal plane.

Computed tomography (CT). Images showing the measurements of the maximum transverse distance of the cleft defect in the sagittal plane.
Statistical analyses were performed using IBM SPSS for Windows (Version 20.0, Armonk, NY). Analysis of variance (ANOVA) was employed to assess significant differences between Veau groups for continuous variables, including age at palatoplasty, time elapsed between surgery and postoperative imaging, and cleft measurements. For statistically significant ANOVA results, multiple posthoc t-tests with Bonferroni correction were conducted to identify specific group differences. To control for potential confounders, including age at palatoplasty and Veau classification, an analysis of covariance (ANCOVA) test was utilized for comparisons of cleft measures between subgroups. A subgroup analysis was performed to compare residual cleft size among patients who underwent palatoplasty using the 2-flap Bardach versus Von Langenbeck techniques. Chi-square tests were conducted to compare categorical variables, such as fistula formation.
A simple linear regression analysis was conducted to examine the relationship between the percentage of the hard palate affected and the rate of fistula formation. The percentage of cleft involvement was used as the independent variable, while the rate of fistula formation served as the dependent variable. The F-statistic, regression coefficient (B), and coefficient of determination (R2) were also calculated to understand model fit, the slope of the regression line, and the proportion of variability in fistula formation that can be explained by percentage of palate with cleft, respectively. Statistical significance was determined using a standard alpha value of .05.
Results
A total of 97 patients had imaging taken more than 1 year after initial palatoplasty at our institution. After excluding 12 patients for alveolar bone grafting, 19 patients for gingivoperioplasties, and 22 patients for orthodontic procedures, 66 patients remained. The majority of patients were white (79%, 52 of 66), with a median age at palatoplasty of 1.02 years (interquartile range: 0.93-1.26 years). The average age at postoperative CT imaging was 13.15 ± 6.61 years, with a mean interval of 11.08 ± 6.89 years between palatoplasty and imaging. Demographic characteristics of all included patients are summarized in Table 1. There were no differences noted between the syndromic versus nonsyndromic patients regarding any demographic values or imaging measures (data not shown).
Patient Demographics and Clinical Characteristics.
Other syndromes included 1 patient with each of the following: CHARGE, Van der Woude, VACTERL association, Stickler, 3MC, facial agenesis, Nager, Rapp Hodgkin, and Opitz.
Abbreviation: IQR, interquartile range.
The majority (84.8%, 56 of 66) of palatoplasties were performed using a 2-flap Bardach technique, while the remaining 15.2% (10 of 66) were performed using the Von Langenbeck technique. CT analysis revealed an average hard palate width of 24.4 ± 5.2 mm, a maximum cleft width of 9.8 ± 3.9 mm in the coronal plane, and a maximum cleft distance of 17.1 ± 8.1 mm across any plane (Table 2). On average, the cleft accounted for 41.4% ± 16.8% of the total hard palate width, measured in the coronal plane. Notably, zero patients demonstrated complete osseous closure of their residual cleft.
Summary of Hard Palate and Cleft Measurements.
Subgroup Analysis by Veau Type
The 3 Veau subgroups (II, III, and IV) were well-matched in terms of age at palatoplasty, time elapsed between surgery and imaging, and hard palate width, as summarized in Table 3.
Comparison of Age at Palatoplasty, Time Between Palatoplasty and Imaging, and Hard Palate Width Between Veau Subgroups.
Preliminary ANCOVA controlling for age at palatoplasty revealed significant differences among the 3 Veau groups in maximum cleft width in the coronal plane (P = .04), maximum cleft distance in any plane (P = .01), and the percentage of total palatal involvement by the cleft (P = .04). Posthoc analysis demonstrated no significant differences in cleft dimensions or the percentage of hard palate involvement between the Veau II and Veau III groups. In contrast, the Veau IV group exhibited significantly larger maximum cleft width, cleft distance, and percentage of cleft palate involvement compared to both the Veau II and Veau III groups, as detailed in Table 4.
Hard Palate and Cleft Measurements Stratified by Veau Type.
Asterisks (*) denote statistical significance (P < .05).
Subgroup Analysis by Palatoplasty Technique
A subgroup analysis was conducted to compare the percentage of hard palate involvement by the cleft between patients who underwent palatoplasty using either the 2-flap Bardach or Von Langenbeck techniques. An ANCOVA was performed, controlling for both age at palatoplasty and Veau subtype. Patients who underwent palatoplasty using the Von Langenbeck technique had smaller residual cleft defects compared to those treated with the 2-flap Bardach technique (Table 5).
Comparison of Residual Cleft Measurements Between 2-Flap Bardach and Von Langenbeck Palatoplasties Controlling for Age at Palatoplasty and Veau Type.
Asterisks (*) denote statistical significance (P < .05).
Subgroup Analysis by Age at Palatoplasty
To assess the impact of age at palatoplasty on the percentage of hard palate involvement by the cleft, patients were stratified into 3 groups based on the timing of surgery: palatoplasty before 1 year, between 1 and 2 years, and after 2 years of age. An ANCOVA was performed controlling for the Veau type. Patients with palatoplasty before age 1 (n = 32) had an average of 26.11% ± 9.24% of the palate involved in the cleft. Patients with palatoplasty between 1 and 2 years (n = 30) had an average of 39.92% ± 11.25% of the palate affected by cleft. Finally, patients with palatoplasty after 2 years of age (n = 4) had an average of 46.35% ± 8.16% of the palate affected. No significant difference in cleft percentage was observed between patients who underwent palatoplasty between ages 1 and 2 compared to those younger than 1 year (P = .24) or older than 2 years (P = .35). However, patients who underwent palatoplasty before age 1 had significantly smaller residual clefts than those who underwent surgery after age 2 (P = .03).
Subgroup Analysis of Fistula Formation by Veau Type
A chi-square analysis comparing fistula formation across all Veau subgroups demonstrated significant differences in the percentage of patients affected by fistula (P = .02). Posthoc analysis revealed that the Veau IV subgroup had a significantly higher percentage of patients with fistula compared to both the Veau II and Veau III subgroups, as detailed in Table 6.
Rate of Fistula Formation by Veau Classification.
Asterisks (*) denote statistical significance (P < .05).
A linear regression model was employed to evaluate the relationship between percentage of the hard palate affected by cleft and the rate of fistula formation. The analysis revealed a significant association between the percentage of the hard palate affected and the rate of fistula formation (P = .01, F = 8.904, B = 0.64, R2 = 0.160). This finding indicates that 16% of the variation in fistula formation could be explained by the percentage of the hard palate affected by the cleft. Furthermore, for every 1% increase in cleft involvement, the rate of fistula formation increased by 0.64%.
Discussion
Among the 66 patients who underwent CT head imaging after palatoplasty, zero achieved complete osseous closure of the hard palate. On average, the residual cleft involved 41.4% ± 16.8% of the hard palate area, with significant variations observed across Veau classifications. When controlling for age at palatoplasty, the time elapsed between palatoplasty and imaging, and hard palate width, subgroup analysis revealed that the Veau IV group exhibited the greatest maximum cleft width, cleft distance, and percentage of palatal involvement. Additionally, the Veau IV group had a significantly higher fistula rate than all other Veau classifications, with regression analysis demonstrating for every 1% increase in hard palate involvement, the fistula formation rate rose by 0.64%. The timing of palatoplasty was also a significant factor, as patients who underwent surgery before age 1 had significantly smaller residual clefts than those who underwent palatoplasty after age 2. Furthermore, the Von Langenbeck technique resulted in smaller residual clefts compared to the 2-flap Bardach technique.
Following primary palatoplasty, zero patients achieved complete osseous palatal closure. These findings are consistent with the existing literature, which has demonstrated incomplete regeneration of the hard palate following soft tissue reconstruction.7-14 Similar to our findings, Choi et al. reported incomplete hard palate osteogenesis in 30 patients who underwent Furlow palatoplasty with a 2-layer mucoperiosteal flap closure. 9 Notably, the postoperative percentage of hard palate involvement in Choi et al.'s study (7.6%) was substantially lower than our cohort (41.4%). This discrepancy likely reflects differences in patient cohorts, surgical techniques, age at palatoplasty, and the exclusion of Veau IV patients. A potential explanation for incomplete palatal osteogenesis lies in the biological limitations of hard palate regeneration. Bone regeneration, irrespective of location, is influenced by several biological factors, including vascular supply, periosteum integrity, and mechanical stress.4,5,7-10 Therefore, disruptions to periosteal growth centers, insufficient vascularization, or excessive mechanical stress can collectively impede full palatal closure.4,5,7-10 Further investigation into the biological constraints of hard palate osteogenesis and their correlation with cleft dimensions is warranted.
This study identified a significant relationship between the timing of palatoplasty and the percentage of hard palate involvement by the cleft defect. Palatoplasty performed before age 1 resulted in significantly smaller residual clefts compared to palatoplasty performed after age 2. These findings contrast with those of Choi et al., who found no significant relationship between the age at palatoplasty and residual cleft dimensions, and Yin et al., who reported the highest rate of bone regeneration (79.4%) in patients undergoing palatoplasty between 4 and 7 years of age.7,9 However, the average timing of palatoplasty in Yin et al.'s cohort (6.1 years old) deviated significantly from the standard practice of operating between 9 and 18 months. 7 As previously stated, bone regeneration is influenced by numerous factors beyond age, and attributing these differences solely to surgical timing may oversimplify a multifactorial process.4,5,7-10 Studies have demonstrated that pediatric bone exhibits a greater density of osteoprogenitor cells and heightened responsiveness to mechanical and biochemical stimuli, facilitating more robust regenerative processes compared to adults.15,16 However, these findings would contradict Yin et al.'s results that patients aged 4 to 7 years should have the highest rates of hard palate osteogenesis. 9 It should be noted that increased regenerative potential is not uniform and can be influenced by systemic factors, further complicating the relationship between age and osteogenesis.4,5,7,15,16 These findings underscore the importance of consistent methodologies when evaluating the impact of surgical timing on hard palate osteogenesis.
Larger cleft defects were associated with higher rates of fistula formation. Regression analysis demonstrated that for every 1% increase in cleft size, the fistula formation rate rose by 0.64%. These findings align with Parwaz et al. and Rossell-Perry et al., which link wider cleft dimensions with an increased risk of fistula formation.12,13 Rossell-Perry et al. proposed that the palatal index—a ratio of cleft width to total palatal width—is a more reliable metric for predicting fistula formation than cleft width alone, as it accounts for the availability of healthy tissue relative to the defect size. 13 Anatomically, larger defects pose significant challenges in achieving tension-free closure, as the need to mobilize and stretch surrounding tissue increases mechanical strain at the repair site.1,2,4,6-8,11 This heightened tension not only elevates the risk of wound dehiscence but also disrupts vascularization, further compounding healing difficulties.4,6,12,13 Given these factors, surgical strategies that minimize closure tension, such as the use of relaxing incisions, local tissue augmentation, or adjunctive bone grafting, may help reduce the risk of postoperative fistula formation and warrant further investigation.
Surgical technique also influenced residual cleft dimensions, as patients who underwent Von Langenbeck palatoplasty had significantly smaller residual defects compared to those treated with the 2-flap Bardach technique. However, this finding must be interpreted with caution, as surgeons preferentially selected Von Langenbeck for narrower clefts, introducing a selection bias. The Von Langenbeck technique, which involves bilateral mucoperiosteal flaps, helps preserve periosteal integrity and vascular supply, potentially creating a more favorable biological environment for osteogenesis.3,17 However, its limited tissue mobilization makes it better suited for narrower clefts, as excessive tension in wider defects can increase the risk of wound dehiscence and fistula formation.1,3,11,18,19 In contrast, the 2-flap Bardach technique, which is more commonly performed at our institution, incorporates lateral relaxing incisions, allowing for greater tissue advancement and tension-free closure.3,20,21 However, this approach necessitates greater periosteal disruption, which may compromise vascularity and limit the recruitment of osteoprogenitor cells.4,13,18 While both techniques utilize soft tissue mobilization to close orofacial defects, each present with distinct advantages and limitations. Given these factors, our findings likely reflect preoperative cleft morphology and surgeon preference rather than the direct effect of surgical technique on osteogenesis. Future studies should incorporate preoperative cleft measurements to further delineate the impact of tissue mobilization strategies on hard palate osteogenesis.
Veau IV patients exhibited the largest cleft palate defects and the highest rates of fistula formation compared to other phenotypes. Bilateral cleft lip and palates are characterized by extensive soft tissue and bony disruption, which generate significant mechanical tension and instability at the repair site. This tension may impair the proliferation and differentiation of osteoprogenitor cells, which are critical for forming stable, vascularized bone.18,22 Additionally, the theory of critical bone regeneration suggests that defects exceeding a specific size threshold lack the biological and mechanical framework necessary to initiate spontaneous healing.7,11,23 Bone grafting has been postulated as a potential solution to address these larger defects, providing structural scaffolding and enhancing the recruitment of osteoprogenitor cells.9,11,23-27 However, the application of bone grafting in hard palate defects has been associated with significant complications, including maxillary hypoplasia due to growth restriction from the graft material.9.10.11,24.25.26 Therefore, the decision to perform bone grafting at the time of primary palatoplasty must balance the potential benefits of early osteogenesis against the risk of long-term maxillary growth restriction. In contrast, patients with Veau II clefts exhibited significantly smaller residual defects and lower rates of fistula formation, reflecting the greater ease of achieving tension-free closure and stable repairs in less severe clefts.
This study has several limitations. The retrospective design introduces inherent selection bias as postoperative CT imaging was not routinely performed for cleft palate evaluation but was obtained incidentally for unrelated clinical indications. Future studies should aim to assess hard palate osteogenesis prospectively, incorporating preoperative morphometric data to validate these findings. Second, the absence of preoperative imaging limits our ability to determine the exact quantity and rate of hard palate osteogenesis. However, this limitation is common among similar studies, as preoperative CT scans are not considered standard of care. Future research should explore more qualitative methods of bone evaluation, such as histology, immunohistochemistry, or PCR. Third, data on postoperative fistula repair strategies were not systematically recorded, limiting our ability to assess how secondary interventions influenced long-term outcomes. Fourth, we excluded patients who underwent alveolar bone grafting, gingivoperioplasty, or orthodontic interventions that may have affected hard palate growth or positioning. While this exclusion criterion aimed to isolate the effects of palatoplasty on hard palate osteogenesis, it also limits the generalizability of our findings, as many cleft patients eventually require additional interventions that could impact bone remodeling. Many patients in our study were lost to follow-up after their initial palatoplasty, resulting in missed opportunities for comprehensive cleft care, including alveolar bone grafting and orthodontic treatment. These patients re-presented many years later, at which point repeat imaging was performed and used for our analysis. Because this cohort did not follow our institution's standard protocols for comprehensive cleft care, the applicability of our findings may be limited when considering patients who do receive indicated secondary cleft procedures in a timely manner. Our study was also limited by the lack of comprehensive fistula classification, which would have added additional insights into fistula location. Finally, this study did not evaluate the specific location of hard palate osteogenesis, which could provide valuable insights into the patterns and variability of bone formation.
Despite these limitations, this study has notable strengths. It represents the largest and most inclusive cohort to date, encompassing all Veau classifications, and employs a rigorous methodology to analyze postoperative CT imaging obtained more than a decade after palatoplasty. These strengths provide valuable long-term insights into the outcomes of cleft repair and hard palate osteogenesis.
Conclusion
Hard palate osteogenesis following primary palatoplasty remains poorly understood, with existing studies reporting conflicting results. In this study of 66 patients, none achieved complete osseous regeneration after palatoplasty. Veau IV patients exhibited the largest residual cleft defects, with increasing cleft proportions demonstrating a positive linear relationship with fistula formation. Given these findings, autologous bone grafting may offer a potential strategy to enhance hard palate osteogenesis and reduce fistula risk. However, its role in primary palatoplasty remains uncertain due to concerns about maxillary growth restriction. Future studies should aim to clarify the balance between the benefits of early bone grafting and its potential impact on craniofacial development.
Supplemental Material
sj-docx-1-cpc-10.1177_10556656251347737 - Supplemental material for Palatal Osteogenesis After Cleft Palatoplasty: Does the Bone Grow?
Supplemental material, sj-docx-1-cpc-10.1177_10556656251347737 for Palatal Osteogenesis After Cleft Palatoplasty: Does the Bone Grow? by Puja Jagasia, Noah Alter, Andrea Hiller, Jennifer Huang, Sumit Pruthi, Stephane Braun, Izabela Galdyn, Michael Golinko and Matthew E. Pontell in The Cleft Palate Craniofacial Journal
Footnotes
Author Contributions
Puja Jagasia: writing, revising, and data analysis; Noah Alter: writing, revising, and data analysis; Andrea Hiller: writing and revising; Jennifer Huang: data collection; Sumit Pruthi: data collection; Stephane Braun: writing and revising; Izabela Galdyn: writing and revising; Michael Golinko: writing and revising; Matthew E. Pontell: writing, revising, and conceptualization.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical Statement
This study was approved by the institutional review board (IRB no. 17111) and a waiver of informed consent was granted. As such, no patients were consented verbally or in writing.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
