Posterior Palatal Expansion via Subnasal Endoscopy (2PENN) for Maxillary Deficiency: A Pilot Study

Abstract

Objective:

Surgically assisted rapid palatal expansion (SARPE) addresses transverse maxillary deficiency, a known contributor to nasal obstruction. The purpose of this study was to assess the feasibility, preliminary outcomes, and safety of posterior palatal expansion via subnasal endoscopy (2PENN), a modified SARPE procedure, aimed at achieving anterior and posterior maxillary expansion.

Methods:

This prospective case series included consecutive adult patients with findings of transverse maxillary deficiency that underwent the 2PENN procedure from 4/2021 to 4/2022. Patients completed pre- and post-operative clinical evaluations, Nasal Obstruction and Septoplasty Effectiveness (NOSE) questionnaires, and computed tomography (CT), with measures including expansion at the level of the posterior nasal spine (PNS), first maxillary inter-molar distance (IMD), and anterior nasal spine (ANS).

Results:

The cohort (N = 20) was middle-aged (39 ± 11 years), predominantly male (80%), and overweight (BMI 28 ± 4 kg/m²). The majority (85%) of patients had sleep breathing issues, of which 10 (59%) had polysomnography-confirmed obstructive sleep apnea (OSA). Full anterior–posterior separation of the mid-palatal suture line was evident on all post-operative CT scans, with mean expansion at the PNS of 3.6 ± 1.3 mm, IMD of 6.1 ± 1.6 mm and ANS of 7.0 ± 1.6 mm (p < 0.001). Following surgery, mean NOSE scores improved from 57 ± 23 to 14 ± 13 (p < 0.001). One patient required maxillary antrostomy for post-operative sinusitis.

Conclusion:

2PENN is an effective and safe technique for achieving both anterior and posterior maxillary expansion in patients with transverse maxillary deficiency. Further study is warranted to better understand the effect of 2PENN in patients with OSA, particularly as it relates to improving pharyngeal patency.

INTRODUCTION

Transverse maxillary deficiency is a craniofacial growth deformity characterized by asymmetric transverse development of the maxilla with respect to the mandible.¹ This has been etiologically implicated in numerous adverse sequalae including orthodontic problems, nasal obstruction, and sleep breathing issues.^1–3 Non-surgical rapid maxillary expansion (RME) is thought to provide ideal maxillary expansion by resulting in a complete, symmetric separation of the mid-palatal suture.^4,5 However, in skeletally mature adult patients with an ossified mid-palatal suture, surgically assisted rapid palatal expansion (SARPE) is required to facilitate transverse maxillary expansion.⁶

Recent evidence has suggested that transverse maxillary deficiency represents a predisposing factor for breathing-related disturbances, including nasal obstruction and obstructive sleep apnea (OSA).^1,7,8 Accordingly, two modified SARPE-techniques have been described with a focus on addressing breathing-related disturbances. Distraction osteogenesis maxillary expansion (DOME) involves a limited LeFort 1 osteotomy and a mid-palatal split at the anterior nasal spine (ANS) via a sublabial incision, with pterygomaxillary disjunction reserved for patient-specific needs.⁹ Endoscopically assisted surgical expansion (EASE) uses minimally-invasive pterygomaxillary disjunction combined with a trans-nasal endoscopic mid-palatal osteotomy to the level of the posterior nasal spine (PNS).¹⁰ DOME and EASE have been shown to increase nasal cavity dimensions, improve subjective nasal breathing, and potentially reduce OSA severity.^9–16

Although these techniques have been shown to improve breathing-related sequalae of transverse maxillary deficiency, limitations remain. Specifically, although DOME has been shown to achieve posterior maxillary expansion, it appears to substantially favor expansion of the anterior maxilla and nasal floor width.¹⁷ Additionally, although EASE provides a minimally-invasive surgical approach to maxillary expansion, it has been both challenging to reproduce from a technical standpoint and create mucosal injury due to reliance on trans-mucosal incisions to complete the mid-palatal osteotomies.^15,16 Given these limitations, there remains a need for a reproducible surgical approach for maxillary expansion that provides reliable posterior maxillary expansion, allows for intraoperative confirmation of a complete mid-palatal osteotomy, and attempts to minimize intranasal trauma.

The purpose of this study was to assess the feasibility and preliminary outcomes of a new modified SARPE procedure incorporating a submucosal osteotomy of the mid-palatal suture under full endoscopic visualization: posterior palatal expansion via subnasal endoscopy (2PENN). The aims of this study were to report (1) measures of anterior and posterior maxillary expansion achieved with 2PENN, (2) subjective nasal breathing outcomes associated with 2PENN, and (3) safety and adverse events related to 2PENN.

MATERIALS AND METHODS

Study Design and Setting

We conducted a retrospective case series of all patients that underwent 2PENN from 4/2021–4/2022 at a single tertiary medical center. Subjects included adult patients (aged 18 years or older) who presented to the sleep surgery clinic of the senior author for surgical management of transverse maxillary deficiency resulting in orthodontic malocclusion, nasal obstruction, and/or sleep breathing issues. This study was approved by the Institutional Review Board at the University of Pennsylvania (IRB 851545).

Study Procedures

Imaging.

All patients underwent non-contrast computed tomography (CT) pre-operatively and after desired maxillary expansion. All pre-operative CT scans were performed after expander placement using the Penn CT SLEEP protocol.¹⁸ Post-expansion CT scans were taken from either the referring orthodontist’s office (cone beam CT) or using the Penn CT SLEEP protocol.

Orthodontics.

Before surgery, all patients underwent application of a maxillary expander by their orthodontist under local anesthesia. There were two orthodontists formally involved in this series (NB and ME). One additional orthodontist supervised the post-operative expansion and orthodontia for one of our patients, in collaboration with one of the co-authors (ME); however, they were not formally involved in this series. Patients were fitted with one of three types of palatal expanders: a bone-borne acrylic expander, a bone-borne metallic maxillary skeletal expander (MSE), or a tooth-borne metallic expander. Selection of maxillary expander type was at the discretion of the orthodontist. Approximately, 5 days after surgery, patients commenced expansion under the supervision of their orthodontist over a 2-6-week period. The rate of expansion is typically about 0.5 mm/day, with some variability based on the expander type used and underlying patient factors. The total goal expansion was determined by their orthodontist, generally 5–10 mm. Following expansion, patients underwent orthodontic realignment of their teeth, as well as retention of the expander on the palate until the palatal bone had healed.

Virtual surgical planning.

Before surgery, a virtual surgical planning (VSP) session was held for every patient using the ProPlan CMF^™ platform in a live interactive session with clinical engineers (DuPuy Synthes, Switzerland). Using the patient’s pre-operative CT scan and 3D modeling software, custom cutting guides were created to optimize patient safety and surgical efficiency (Fig. 1).

Fig. 1.

Virtual surgical planning (VSP) custom titanium 3D-printed cutting guides, designed from the patient’s pre-operative CT scan. The anterolateral maxillary osteotomies (a) are designed to clear the tooth roots by at least 5 mm (red dotted arrow), avoid the infraorbital vascular bundles (red dotted circle), and remove a small wedge of bone laterally (w) to prevent bony interferences during expansion. The midline notch (b) identifies the midline between the central incisor tooth roots, with the inferior aspect of the notch determining a 1 mm inter-incisal gap, to ensure that the mid-palatal sagittal split osteotomy avoids the central incisor tooth roots. The guide is designed with an “apron” contour (c, red dotted curve) to minimize soft tissue elevation and with screw holes (*) to stabilize the cutting guide intraoperatively. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

Surgical technique: 2PENN.

The 2PENN technique involves modified LeFort 1 osteotomies with submucosal endoscopic-assisted osteotomy of the mid-palatal suture. Through a sublabial incision, the anterior maxillary face is exposed broadly from lateral buttress to lateral buttress. The mucosa is then elevated from the nasal floor as far posteriorly as possible. The custom cutting guide is introduced to the field and affixed with 2 screws. The anterior and lateral maxillary osteotomies are carried out using piezo-electric saws (Piezosurgery^® by Mectron, Italy). The mid-palatal osteotomy is also initiated anteriorly with the cutting guide. The cutting guide is removed, and the lateral and pterygoid osteotomies are completed. After reduction of the ANS, a supraspinal septal osteotomy is performed, with elevation of the nasal mucosa under visualization with a zero-degree endoscope. Inferior transection of the vomer to its posterior extent marks the termination of the septal osteotomy, confirmed with digital palpation via the oral cavity. Using the inter-incisal midline groove, a piezo-electric saw (UNIVR Insert Tip, #03600008) is used to initiate the mid-palatal osteotomy along the anterior face of the maxilla, extending superiorly to the remnant ANS. Under visualization with a 30-degree endoscope, a longer piezo-electric saw (MT8–20L Insert Tip, #03600013) is used to penetrate the cortex of the roof of the maxilla for the complete mid-palatal osteotomy from PNS to ANS. Simultaneous digital palpation in the oral cavity by the surgeon’s non-operative hand allows for tactile feedback of complete transection of the mid-palatal suture. Osteotomes are then used anteriorly to complete the release of the mid-palatal suture, resulting in a diastema between the central incisors. An osteotome is placed in each of the 5 buttress regions (mid-palatal, bilateral nasomaxillary, and bilateral zygomaticomaxillary) to ensure adequate mobility. Following this “five-point inspection,” the expander is then turned, without resistance, several times to create a diastema of ~1–2 mm, with endoscopic confirmation of PNS expansion. See Figure 2 for surgical highlights and Video S1 for a full summary of the surgical steps.

Fig. 2.

Intraoperative images highlighting key components of 2PENN procedure. (A) Pre-operative nasal endoscopy. (B) Endoscopic visualization of elevated nasal mucosa from floor of nose, demonstrating maxillary crest and the posterior nasal spine (PNS). Note that for this patient, the temporary anchorage devices (TADs) from the maxillary skeletal expander (MSE) were safely visualized (arrows), preventing damage to the MSE during the subsequent mid-palatal sagittal split osteotomy. (C) Endoscopic visualization of mid-palatal sagittal split osteotomy using the piezo-electric saw, with safe visualization and avoidance of the TADs (arrows). (D) Immediate post-operative nasal endoscopy demonstrating intact nasal mucosa. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

Post-operative management.

All patients were offered same day discharge or 23-h post-operative observation, based on patient preference. Patients were placed on a soft diet for a period of 1–2 weeks and discharged with oral analgesics, viscous topical lidocaine, and chlorhexidine oral rinses. The first post-operative visit occurred around 5–7 days after surgery, after which the patients were cleared to commence with maxillary expansion. The second post-operative visit occurred after patients had completed maxillary expansion per their orthodontist.

Data Collection

Data were collected from clinic notes, operative reports, and CT scans. Data on demographics, indications for 2PENN, and past surgical treatments were collected from pre-operative clinic notes. Data on surgical procedures performed concurrently with 2PENN were collected from the operative record. Nasal Obstruction Symptom Evaluation (NOSE) questionnaires were completed during the pre-operative clinical evaluation and after completion of maxillary expansion, typically about 3–4 months after surgery.¹⁹ Polysomnography data were collected from sleep study reports. Pre-operative sleep studies occurred within a 2-year period prior to the 2PENN procedure and included diagnostic polysomnograms, splitnight polysomnograms (diagnostic portion), and home sleep apnea tests, depending on the clinical work up patients received for sleep breathing issues. Post-operative sleep studies occurred following completion of full maxillary expansion, approximately 4–6 months after 2PENN, and included diagnostic polysomnograms and home sleep apnea tests. Data on adverse events or side effects were collected from post-operative clinic notes. Serious adverse event was defined in accordance with the United States Food and Drug Administration.²⁰

Measurements from the pre-operative CT scans and the 3-month post-expansion CT scans were made using the software Invivo 6 (Anatomage, San Jose, California). Prior to performing measurements, all scans were placed into standard alignment controlling for yaw, pitch, and roll. The following pre- and post-operative measurements were made in the coronal plane: (1) PNS expansion (post-operative CT only); (2) greater palatine foramina distance (GPFD – distance between the greater palatine foramina at turn from vertical to horizontal); (3) inter-molar distance (IMD – distance between the bony alveolar edges of the first molars at the level of the furcation); (4) inter-molar height (IMH – midline hard palate to perpendicular formed by plane of IMD); (5) ANS expansion (post-operative CT only); (6) piriform width (PW – measured from level of inferior turbinate attachment and internal nasal valve). The following measurements were made in the sagittal plane: (1) distance between most posterior functional portion expander and PNS (pre-operative CT); (2) distance from ANS to PNS (pre-operative CT). Measurements were taken by two study team members (SMJ and ERT), with joint arbitration for measures discrepant by >1.0 mm (Fig. 3).

Fig. 3.

Representative coronal CT scans (left panels) used for anatomic measures of expansion, with corresponding anteroposterior location (red circle) demonstrated on respective axial CT scan (right panels): (A) PNS (red horizontal line), (B) IMD (red horizontal line) and IMH (red vertical line), (C) ANS (red horizontal line). [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

Statistical Analysis

Descriptive data were reported as mean ± standard deviation for continuous variables and as numbers and percentages for categorical variables. Univariate analyses comparing outcomes (CT measures, NOSE scores) pre- and post-operatively were performed using the paired t-test. Exploratory analyses testing the association between expander type and PNS expansion, the correlation between posterior positioning of the expander and PNS expansion, and post-operative change in apnea-hypopnea index (AHI) were performed using analysis of variance (ANOVA), Pearson correlation, and Student’s t-test, respectively. A two-tailed p-value < 0.05 was considered statistically significant. All analyses were performed with Stata/SE 14 (StataCorp LP, College Station, Texas).

RESULTS

Baseline Characteristics

Twenty patients underwent 2PENN for treatment of transverse maxillary deficiency (Table I). On average, the cohort was middle-aged (39 ± 11 years), predominantly male (80%), predominantly Caucasian (70%), and overweight (BMI 28 ± 4 kg/m²). Seventeen patients (85%) had sleep breathing issues on evaluation, of which 10 (59%) had polysomnography-confirmed OSA. Many patients had multiple sequelae of transverse maxillary deficiency (Table I). Most patients had acrylic expanders placed by their orthodontist.

TABLE I.

Patient Characteristics.

Characteristic	All Patients (N = 20)
Age, mean ± SD, years	39 ± 11
Men, no. (%)	16 (80)
White race, no. (%)	14 (70)
BMI, mean ± SD, kg/m2	28 ± 4
Past history of nasal surgery, no. (%)	6 (30)
Indications for surgical expansion, no. (%)
Sleep breathing issues	17 (85)
Malocclusion	14 (70)
Nasal obstruction	13 (65)
Type of maxillary expander, no. (%)
Acrylic (bone-borne)	14 (70)
Maxillary skeletal expander (bone-borne)	3 (15)
Tooth-borne	3 (15)

BMI = body mass index.

Anatomic Effects of 2PENN

Table II provides a summary of anatomic measures assessed by CT, before and after 2PENN with completion of supervised maxillary expansion. All patients achieved complete anterior–posterior split of the mid-palatal suture; however, one patient had an asymmetric split, resulting in no expansion at the PNS. Mean expansion at the PNS, IMD, and ANS was 3.6 ± 1.3 mm, 6.1 ± 1.6 mm, and 7.0 ± 1.6 mm, respectively (p < 0.001). On average, this transverse expansion was associated with a reduction in the height of the maxilla (mean IMH reduction −1.4 ± 1.4 mm, p < 0.001). Mean expansion at the GPFD, and additional measure of posterior expansion, was 2.9 ± 1.1 mm (p < 0.001). Of note, mean PW expansion, the only measure of expansion superior to the osteotomies, was 1.9 ± 1.0 mm (p < 0.001).

TABLE II.

Anatomic Measures: Pre- & Post-2PENN Procedure & Expansion (N = 20).

Measurement	Pre-2PENN, Mean ± SD (mm)	Post-2PENN, Mean ± SD (mm)	Change	p-Value
PNS	–	3.6 ± 1.3	+3.6 ± 1.3	<0.001
IMD	30.0 ± 3.5	36.1 ± 3.1	+6.1 ± 1.6	<0.001
IMH	9.7 ± 2.1	8.3 ± 1.9	−1.4 ± 1.4	<0.001
ANS	–	7.0 ± 1.6	+7.0 ± 1.6	<0.001

ANS = anterior nasal spine; IMD = first inter-molar distance; IMH = height at IMD horizontal; PNS = posterior nasal spine; PW = piriform width.

Changes in Subjective Nasal Obstruction

A mean reduction in NOSE scores was also noted pre- and post-2PENN with maxillary expansion (57 ± 23 to 14 ± 13, p < 0.001). Of note, 11 of the 20 patients (55%) had an additional nasal-related source of anatomic obstruction (i.e., septal deviation, turbinate hypertrophy, etc.) and underwent concurrent nasal surgery (septoplasty, turbinate reduction, etc.) at the time of their 2PENN. In patients undergoing 2PENN alone, the mean NOSE score reduction was 37 ± 17. In patients undergoing concurrent nasal surgery, the mean NOSE score reduction was 46 ± 27. The change in NOSE scores between these two groups was not significantly different (p = 0.43). Six of the 20 patients (30%) had previously undergone at least one nasal procedure in their lifetime prior to undergoing 2PENN.

Exploratory Analyses

With regards to PNS expansion, the degree of expansion was neither associated with expander type (p = 0.24) nor correlated with posterior placement on the maxilla (p = 0.46). In regard to reduction AHI, six patients with mild–moderate OSA underwent 2PENN as a standalone procedure for OSA (Table III). In these patients, mean AHI pre-expansion was 15 ± 12 events/h and post-expansion was 3 ± 2 events/h (p = 0.06).

TABLE III.

Patient Characteristics – OSA Patients (n = 6).

Characteristic	OSA Patients (n = 6)
Age, mean ± SD, years	48 ± 13
Men, no. (%)	3 (50)
White race, no. (%)	5 (83)
BMI, mean ± SD, kg/m2	30 ± 4
Epworth sleepiness scale (0–24)*	12 ± 6
Apnea-hypopnea index (events/h)†	15 ± 12
Past history of nasal surgery, no. (%)	2 (33)
Indications for surgical expansion, no. (%)
Sleep breathing issues	6 (100)
Malocclusion	3 (50)
Nasal obstruction	4 (67)
Drug-induced sleep endoscopy collapse pattern
Complete antero-posterior collapse at velum, no. (%)	4 (67)
Complete antero-posterior collapse at tongue base, no. (%)	4 (67)

BMI = body mass index.

^*Pre-operative, available in 5 (83%) of patients.

^†Pre-operative, available in 5 (83%) of patients.

Adverse Events

One serious adverse event was noted in the post-operative period. One patient with a history of recurrent acute sinusitis developed post-surgical sinusitis. This did not resolve with medical management and required operative maxillary antrostomy to relieve the sinusitis 1 month after 2PENN. Additional sequelae experienced by individual patients included: persistent hypoesthesia of upper teeth; temporary maxillary hypermobility; intra-operative screw fracture during removal of VSP guide (practice because changed to pre-drilling fixation holes); patient concern regarding cosmesis of the nasal base (without affecting nasal breathing), asymmetric maxillary expansion (resolved with orthodontics), lip abrasion secondary to use of a conventional drill (practice because changed to piezo-electric technology), palatal granuloma (resolved at 3 months) and stitch pseudo-abscess (resolved with suture removal). The patient that experienced asymmetric maxillary expansion resulted from incomplete release of the nasomaxillary buttress on one side and led to the practice of performing the “five-point inspection” as a routine step in the 2PENN procedure to confirm mobility in all 5 buttress regions. None of the patients in this series were found to have a de novo septal perforation at their 3-month follow up visit (two of the patients had pre-existing septal perforations from prior nasal surgeries, which were stable in appearance).

DISCUSSION

The results of this pilot study demonstrate that 2PENN consistently and safely achieved maxillary expansion. Specifically, a complete anterior–posterior split of the mid-palatal suture was radiographically confirmed in all patients, with reproducible posterior maxillary expansion. Additionally, all patients experienced a subjective improvement in nasal breathing following 2PENN, even when performed in the absence of a concurrent nasal procedure. Finally, one serious adverse event requiring additional surgical intervention (maxillary antrostomy to relieve post-surgical sinusitis) was reported in the 90-day post-operative period.

Although other SARPE procedures have been shown to successfully address manifestation of transverse maxillary deficiency, each surgical variant of this procedure has its own limitations. Both traditional SARPE and DOME employ an anterior maxillary approach for achieving a midline split of the mid-palatal suture and rely on a blind mid-palatal split which terminates in the anterior aspect of the maxilla and does extend to the PNS.^6,9–14,17 Previous studies have demonstrated that the posterior region of the maxilla is particularly challenging to expand, even with surgical assistance. This may result in preferential anterior widening with less reliable expansion of the posterior nasal floor.^21–23 DOME, which reserves pterygomaxillary disjunction for patient-specific needs, may be particularly suscepti-ble to this discrepancy as separation of the pterygomaxillary suture has been shown to be important for achieving posterior maxillary expansion.^24,25 In contrast, EASE employs strategic separation of the pterygomaxillary suture in combination with trans-nasal endoscopic, bilateral transmucosal mid-palatal osteotomies posteriorly to the level of the PNS, with the goal of facilitating posterior maxillary expansion.^15,16 Indeed, EASE has been shown to achieve posterior maxillary expansion, with associated expansion of the posterior nasal floor. However, reliance on transmucosal osteotomies inherently results in increased intranasal mucosal trauma, with potential for increased intraoperative bleeding, reduced endoscopic visualization, and for post-operative epistaxis. As such, this approach has been technically challenging to reproduce and has had limited adoption in contemporary clinical practice.

2PENN has several technical advantages which address these limitations. Although the other modified SARPE procedures include endoscopic elements both DOME and EASE utilize endoscopy for portions of their respective procedures, a unique aspect of the 2PENN is the submucosal endoscopic-assisted osteotomy of the mid-palatal suture. Elevation of the nasal mucosa from the nasal floor under endoscopic visualization allows for an atraumatic, broad exposure of the roof of the maxilla, including the entire mid-palatal suture from PNS to ANS (Fig. 2). This allows for complete visualization when performing the mid-palatal osteotomy, providing intraoperative confirmation of a complete anterior–posterior split of the mid-palatal suture, confirmation of a symmetric separation of the maxilla, and avoidance of damage to the temporary anchorage devices (TADs) of a patient’s maxillary expander (Fig. 2). Moreover, the use of an endoscope to achieve this visualization leverages the skill set of the otolaryngologist, which may help facilitate broader adoption of this modified SARPE procedure by otolaryngologists. An additional feature is the routine use of VSP for this procedure. This allows for careful anticipation and placement of the osteotomies in a manner which increases the predictability of the procedure and thus optimizes surgical efficiency. It also incorporates key surgical landmarks, including the V2 neurovascular bundles and the location of the dental roots, which reduces the risk of neurovascular and dental complications, and thus further enhances surgical efficiency (Fig. 1). Lastly, it allows for the identification of anatomic variants—such as maxillary asymmetry and the sequalae of previous craniofacial trauma—that could potentially make surgery more challenging and increase the risk of asymmetric expansion in the post-operative period.

In this study, there was variability in reported measures of maxillary expansion, with measures of posterior expansion similar to those reported in other airway-oriented SARPE procedures.^15–17 Like with most SARPE procedures, patients still demonstrated greater anterior expansion in a fan-like pattern, as demonstrated by the measures of expansion. PNS expansion is prone to some variability, given the separation resulting from corresponding bone removal and separation from the osteotomy, itself. Notably, GPFD expansion, which is not prone to this variability, demonstrated similar magnitudes of expansion, which suggests consistency in achieving posterior expansion from the 2PENN. Interestingly, the PW also demonstrated reliable expansion despite being the only measure taken superior to the mobilized maxillary segments resulting from the modified LeFort 1 osteotomies. Our observed expansion measures for the PW thus also represent expansion occurring in non-mobilized portions of the maxilla, which may account for the improvement in NOSE scores documented by patients in this cohort and is comparable to a measure associated with clinical efficacy recently reported in a cohort of patients who underwent DOME.¹⁷

One important source for variability in measures of maxillary expansion is the collaborating orthodontist. The orthodontist ultimately decides how much post-operative expansion is necessary, based on the orthodontia required to achieve proper alignment of the teeth. The orthodontist also oversees several important considerations including maxillary expander type and positioning, both of which may affect degree of posterior maxillary expansion. On exploratory sub-analysis, there was no association between expander type, or more posterior maxillary expander placement, and PNS expansion. Because we predict that these maxillary expander characteristics may affect posterior maxillary expansion, cohorts with larger sample sizes are needed to further test the effects of these characteristics.

A strength of this study was its use of imaging (Figs 3 and 4). All patients underwent a pre- and post-expansion CT scans with standardized measurements taken and corroborated by two research team members to minimize variability in measurements. Another strength of this study was use of the validated NOSE questionnaire to measure subjective nasal obstruction pre- and post-expansion. Furthermore, all patients had at least 3 months of follow up after the 2PENN procedure, with corresponding post-expansion CT scans and NOSE scores available for pre-operative comparison. Lastly, all patients underwent 2PENN with the same surgeon, ensuring consistency in surgical technique throughout the study.

Fig. 4.

Pre- and post-expansion CT scans of a patient with 3D reconstructions of the maxilla: (A) anterior view, and (B) superior view. Note a complete split was achieved from ANS to PNS, with expansion tapered in an anterior to posterior direction. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]

There are important limitations to this study. As a pilot case series, this study has no comparison group for assessment of measures of expansion. Additionally, there was significant variability in measures of maxillary expansion noted on CT. We attempted to mitigate this through standardization of CT acquisition and expansion measures, as well as corroboration of these measures with two research team members. Patients also exhibited various, seemingly diverse presenting complaints of transverse maxillary deficiency, inherently introducing heterogeneity to the data. Because of the diversity of presenting complaints, some of the patients had previous treatment for nasal obstruction and others pre-existing OSA, further introducing heterogeneity and potential confounding to this study. Curiously, even patients presenting initially with malocclusion as their primary complaint, without necessarily complaining of nasal obstruction, had NOSE scores greater than 40, suggesting that occult nasal obstruction may be prevalent in patients with transverse maxillary deficiency. It is also important to note that over half the patients in this cohort had a nasal procedure concurrent to 2PENN, because of a nasal-related source of anatomic obstruction, and thus the effect on nasal breathing in these patients was not solely from 2PENN. However, reductions in post-expansion NOSE scores in patients that underwent 2PENN alone were dramatic and similar to patients who underwent 2PENN with concurrent nasal surgery, suggesting that 2PENN may have an important independent effect on nasal breathing. Future studies will incorporate additional validated measures of nasal breathing, in addition to the NOSE questionnaire. Finally, although not statistically significant nor the focus of this pilot study, the marked reduction in AHI suggests a potentially clinically important improvement in OSA with 2PENN. We hypothesize that posterior maxillary expansion, as is achieved by 2PENN, may play an important role in improving pharyngeal patency in select patients with OSA. We plan to assess this further in future, larger-scale controlled studies that more rigorously assess anatomic and patient-reported outcome measures following the 2PENN procedure.

CONCLUSION

In this pilot study, 2PENN effectively and safely achieved maxillary expansion, including posterior maxillary expansion. Ideal surgical candidates include patients with transverse maxillary deficiency resulting in orthodontic malocclusion, nasal obstruction, and/or sleep breathing issues. Further study is warranted to assess the effect of posterior maxillary expansion on pharyngeal patency in patients with OSA.