Influence of second premolar extractions on the volume of the oral cavity proper: a control comparative cone-beam computed tomography volumetric analysis study

ABSTRACT

Objectives

To compare the pre- and postorthodontic treatment volumetric changes of the oral cavity proper (OCP) in extraction and nonextraction patients and to identify influencing variables.

Materials and Methods

Pre- and posttreatment cone-beam computed tomography (CBCT) scans of patients undergoing orthodontic treatment with fixed labial appliances and who satisfied the inclusion criteria were individually landmarked. Linear, angular, and volumetric measurements were determined. Descriptive statistics, repeated measure analyses of variance, correlations, and stepwise regression statistical analyses were applied.

Results

The CBCT scans of 54 patients who underwent the extraction of second premolars and/or extraction of second primary molars associated with agenic second premolars, and 59 nonextraction patients matched for crowding, were assessed. The mean age for both groups was 15 years. There was a statistically significant increase in the volume of the OCP for both the extraction and nonextraction groups, with the nonextraction group demonstrating a larger increase in the volume of the OCP. Gender, age, changes in mandibular and maxillary arch length, and changes in mandibular and maxillary intermolar width all influenced the change in the OCP volume.

Conclusions

The volume of the OCP increased in growing patients with and without the extraction of the second premolars and/or extraction of second primary molars associated with agenic second premolars. Patients who did not have extractions as part of their orthodontic treatment demonstrated a greater overall increase in OCP volume.

INTRODUCTION

Notwithstanding the difficulty in accurately defining and measuring the oral cavity, the mouth can be broadly divided into two parts: the vestibule and the oral cavity proper (OCP).¹ The OCP lies internal to the maxillary and mandibular dental arches. The superior border is formed by the hard and soft palates, whereas the inferior border is formed by the tongue and the soft tissue structures of the floor of the mouth.¹ It is separated from the oropharynx via a ring of structures that include the soft palate, anterior tonsillar pillars, the pterygomandibular raphe, and posterior portion of the tongue.^1,2 The terms intraoral, oral cavity, and OCP are frequently used interchangeably. However, the term OCP will be used in this article.

Measurement of the size and volume of the OCP has been carried out using a variety of methodologies including alginate and polyvinylsiloxane impressions, lateral cephalometry, computed tomography (CT) scans, and magnetic resonance imaging scanning.^3–5 Although these have provided valuable information, potential disadvantages are patient discomfort and, often, high radiation dosage. The advent of cone-beam computed tomography (CBCT) has created an opportunity for three-dimensional assessment of the OCP, with the ability to obtain more accurate linear and volumetric measurements while keeping costs and radiation exposure at acceptable levels.^4,6

A large tongue in a normally sized OCP or a normal tongue in a smaller OCP has been linked to malocclusions, as observed in Beckwith-Wiedemann syndrome in which tongue enlargement results in lateral open bite and proclination of the lower dentition.⁷ It has also been hypothesized that a mismatch between tongue size and OCP creates an environment in which the tongue is obliged to rest posteriorly, which may result in restriction of the airway and consequent predisposition for obstructive sleep apnea (OSA).^2,7

Extraction of teeth as part of an orthodontic treatment plan has long been practiced and remains controversial.⁸ The debate regarding the impact of orthodontically prescribed extractions on temporomandibular joint dysfunction and dental and facial esthetics is ongoing.^9–12 Several researchers have indicated that a new concern has emerged that contends that shortening of the dental arch length after orthodontic treatment by premolar extractions results in a reduction of the OCP volume.^13,14 This may result in insufficient space for the tongue, which may subsequently negatively influence the airway. It is essential, therefore, that orthodontists understand the potential clinical consequences of changes in OCP volume as a result of orthodontic treatment.

The volumetric changes to the OCP as a result of orthodontic extractions has not been previously investigated. The aim of this study was to compare the pre- and postorthodontic treatment volumetric changes of the OCP in a sample of patients undergoing extraction and nonextraction orthodontic treatment. The null hypothesis was that the volume of the OCP would not change as a result of extractions prescribed as part of orthodontic treatment.

MATERIALS AND METHODS

Ethical approval for the investigation was provided by the University of Adelaide. A power study indicated that a sample size of 37 was required to determine significance in a mean difference of 3000 mm³ in OCP volume between pretreatment (T0) and posttreatment (T1).

The database of a private orthodontic practice in Adelaide, South Australia, was extensively audited for patients who had undergone comprehensive orthodontic treatment between 2015 and 2021. The inclusion criteria for selection were the following:

• no history of previous orthodontic treatment,

• complete T0 and T1 CBCT scans with appropriate extensions,

• complete patient records,

• no craniofacial deformities, and

• single-phase treatment using Tip-Edge (TP Orthodontics, La Porte, Ind.) full-fixed appliances.

In addition, patients in the experimental group had undergone extraction of both maxillary second premolars or all second premolars or had agenic second premolars in which the extraction of primary second molars was required in addition to the contralateral primary second molar or second premolar.

The control group was selected to match the level of crowding of the experimental group. Both groups were divided into two subgroups according to the level of T0 crowding: mild or moderate/severe.

CBCT images of the patients were obtained using a Carestream Health CS9300 scanner; Carestream Health, Rochester, N.Y.), with each voxel measuring 0.3 mm × 0.3 mm × 0.3 mm and a scan time of 12–20 seconds. The standardized image acquisition protocol required patients to breathe in and position the tongue on the palate during exposure. Acquired data sets had approximately 442 images saved in Digital Imaging and Communications in Medicine format. The images were first cleaned of noise by a combination of blurring, cropping, and ray-tracing the affected regions to reconstruct the density values that were deemed to interfere with the smoothness required to observe the landmarks and perform calculations.

The landmarks for evaluation were identified and placed manually by Dr Mladenovic (Table 1).¹⁵ A screen capture of the landmarking program is shown in Figure 1.

Table 1.

Landmarks Used

Upper left and right central incisor

Lower left and right central incisor

Upper left and right first permanent molar

Lower left and right first permanent molar

Lingula

Hamulus notch left and right

Menton

A point

B point

Gonion left and right

Anterior nasal spine

Posterior nasal spine

Sella

Nasion

Figure 1.

Screen capture image of the landmarking program.

Once landmarking was completed, the CBCT slices were processed through a customized software program for linear and volumetric calculations (Figure 1). The CBCT scans were oriented to a standardized position via the anterior nasal spine, posterior nasal spine, and menton landmarks. The landmarks were connected via computerized bridging to generate continuous anteroposterior and lateral boundaries of the OCP. Superior and inferior boundaries were set via automated isolation of the hard tissues using the hard palate and inferior border of the mandible (Table 2, Figure 2).

Table 2.

Boundaries of the Intraoral Cavity Proper^a

Anterior boundary	Bridging the palatal vault to the anterior incisor landmarks down to menton
Lateral boundary	Bridging the maxillary incisors landmarks to the posterior molar landmarks of the upper and lower dentition
Superior boundary	Palatal vault isolation bridging between along with ANS-PNS
Posterior boundary	Bridging the PNS landmark with the left and right lingula, the lingula was then bridged to the gonion landmarks
Inferior boundary	Outline of inferior surface of the mandible bridged to the apex of incisors anteriorly and lingual from gonion to posteriorly

^a ANS indicates anterior nasal spine; PNS, posterior nasal spine.

Figure 2.

Example of the two-dimensional reconstruction image of boundaries selected for calculation.

Linear and angular measurements were collated as listed and defined in Table 3.

Table 3.

Linear and Angular Definitions^a

Max AL	Linear measurement from a fixed vertical plane at PNS perp to the ANS-PNS plane to the 11 Inc edge
Man AL	Linear measurement from a fixed vertical plane at PNS that is perpendicular to the ANS-PNS plane to the 31 Inc edge
Max IMW	Measurement from the MB cusp of 16 to 26
Man IMW	Measurement from the MB cusp of 36 to 46
Man Inc Ang	Ang of 31 in relation to the Man plane
Max Inc Ang	Ang of 11 in relation to ANS-PNS plane
Inter Gonial Dist	Dist between the left and right Gonion landmarks
Max–Man angle	Ang of the ANS-PNS plane with Man plane angle
ANB angle	Difference of angles SNA and SNB

^a AL indicates arch length; Ang, angulation; ANS, anterior nasal spine; Dist, distance; IMW, intermolar width; Inc, incisor; Man, mandibular; Max, maxillary; MB, mesiobuccal; and PNS, posterior nasal spine. SNA, Angle formed by the intersection of sella-nasion and nasion-A lines; SNB, Angle formed by the intersection of sella-nasion and nasion-B lines.

Following completion of landmarking, a voxel sensitivity threshold was developed via a K-means clustering algorithm. This enabled differentiation of the soft tissues, air, and hard tissue voxel radiodensities. The isolation of voxels was then used for the calculation of the volume of the area defined by the landmarks. The volume was calculated in cubic millimeters and was generated in a .text format by the program.

Data were recorded on an Excel spreadsheet (Microsoft Corp, Redmond, Wash). Descriptive statistics were calculated in means and percentages. Continuous variables were compared using t-tests, and categorical variables were compared using the chi-square test to determine whether the groups were comparable at baseline. The Shapiro-Wilk test was applied to test for normality. The underlying assumptions for the parametric tests were examined for each analysis. No adjustments were required for the normalization of the data. The normality assumptions were confirmed via the homogeneity of variance and Levene’s test.

Repeated measures analyses of variance tests were performed to test for differences across time and between groups (extraction vs nonextraction). The significance was set at P ≤ .05.

Correlation testing was used to assess for linear relationships to explore potential influences on the change in intraoral volume. This was followed by regression and multiple regression analyses to identify confounders. Statistical analyses were carried out using SPSS statistics software version 27 (IBM, Armonk N.Y.).

RESULTS

Application of inclusion criteria resulted in 54 patients in the experimental (extraction) group and 59 patients in the control (nonextraction) group. Table 4 shows that there were no significant differences in age and gender between the groups.

Table 4.

Descriptive Statistics of the Extraction and Nonextraction Groups (N = 113)

Variable	Nonextraction, n = 59	Extraction, n = 54	P Value
Sex, n (%)			.42
Male	24 (41)	18 (33)
Female	35 (59)	36 (67)
Age, y
Mean age at T0	15.0	15.0	.743
≤15, n (%)	48 (81)	43 (80)	.817
≥16, n (%)	11 (19)	11(20)
Skeletal classification, n (%)
Class I	42 (71)	21 (39)	.001
Class II	17 (29)	33 (61)
Initial crowding, n (%)
Spacing/mild, ≤3 mm	25 (42)	25 (46)	.675
Moderate/severe, >3 mm	34 (58)	29 (54)
Maxillo-mandibular angle, n (%)
≤24	38 (64)	26 (48)	.102
25–31	18 (31)	25 (46)
≥32	3 (5)	3 (6)
Extraction, n (%)
Upper second premolars	N/A	20 (37)	N/A
Four second premolars	N/A	34 (63)
Average treatment length, months, n (%)	21.5 (7.0)	23.5 (6.4)	.11

^a N/A indicates not applicable; T0, pretreatment.

Figure 3 shows that there was a significant (P = .018) increase in the OCP volume change in the extraction and nonextraction groups over time, with the nonextraction group increasing by a comparatively greater amount than the extraction group (F_1,53 = 12.0, P = .001).

Figure 3.

Pretreatment (T0) and posttreatment (T1) changes in the oral cavity proper volume.

Figure 4 shows that, regardless of the level of crowding, OCP volume increased at the same rate over time. There was no significant difference in the rate of volume increase between patients with different levels of crowding (F_1,111 = 0.25, P = .875).

Figure 4.

Pretreatment (T0) and posttreatment (T1) changes in the oral cavity proper volume based on the level of crowding.

Table 5 shows the association of variables with changes in OCP, and Table 6 illustrates that there was no difference in the linear measurements between the extraction and nonextraction groups at T0, but statistically significant differences were present at T1 and in the differences between T0 and T1 (P ≤ .05). Table 7 shows that 11.9% of the difference in the change in OCP volume was explained by gender.

Table 5.

Association of Variables With Changes in OCP

	r	P Value
Age	−0.259	.006
Max AL	0.235	.012
Man AL	0.396	<.001
Max IMW	0.278	.012
Man IMW	0.248	.008

^a AL indicates arch length; IMW, intermolar width; Man, mandibular; Max, maxillary; and OCP, oral cavity proper.

Table 6.

Comparison of Linear Measurements Between the Extraction (n = 54) and Nonextraction (n = 59) Groups^a

Variable	Mean (SD), mm	P Value*
Max AL (T0)
Nonextraction	49.98 (4.77)	.66
Experimental	50.47 (7.03)
Max AL (T1)
Nonextraction	50.57 (4.10)	.001
Extraction	47.68 (6.03)
Max AL difference (T0–T1)
Nonextraction	0.59 (2.64)	<.001
Extraction	−2.79 (5.54)
Man AL (T0)
Nonextraction	46.11 (4.39)	.74
Extraction	46.48 (7.00)
Man AL (T1)
Nonextraction	48.30 (3.87)	.01
Extraction	45.54 (5.91)
Man AL difference (T0–T1)
Nonextraction	2.19 (2.08)	<.001
Extraction	−0.94 (5.04)
Max IMW (T0)
Nonextraction	39.71 (2.50)	.45
Extraction	39.31 (3.15)
Max IMW (T1)
Nonextraction	39.63 (5.00)	.001
Extraction	37.20 (3.24)
Max IMW difference (T0–T1)
Nonextraction	−0.08 (5.29)	.01
Extraction	−2.11 (2.55)
Man IMW (T0)
Nonextraction	44.55 (2.53)	.50
Extraction	44.22 (2.75)
Man IMW (T1)
Nonextraction	44.25 (2.14)	<.001
Extraction	41.49 (3.19)
Man IMW difference (T0–T1)
Nonextraction	−0.31 (2.11)	<.001
Extraction	−2.73 (2.69)

^a AL indicates arch length; IMW, intermolar width; Man, mandibular; Max, maxillary; SD, standard deviation; T0, pretreatment; and T1, posttreatment.

Table 7.

Regression Analysis Showing the Impact of Extraction, Gender, Age, Max AL, Man AL, Max IMW, and Man IMW on the Volume of OCP^a

Variable	Model	Adjusted R2		UnstandardizedCoefficients		StandardizedCoefficients	t	Significance
				B	Std. Error	β
Extraction	1	0.041	(Constant)	8118.0	1096.46		7.40	.000
Extraction	−3814.45	1586.11	−0.22	−2.41	.018
Gender	1	.119	(Constant)	11189.34	1406.79		7.95	.000
			Extraction	−3434.19	1524.89	−0.20	−2.25	.026
Gender	−5177.38	1576.18	−0.29	−3.28	.001
Age	1	.104	(Constant)	15223.05	2614.39		5.82	.000
			Extraction	−3956.61	1533.67	−0.23	−2.58	.011
Age	−474.74	159.69	−0.27	−2.97	.004
Max AL	1	.060	(Constant)	7921.454	1091.064		7.260	.000
			Extraction	−2690.317	1690.129	−0.157	−1.592	.114
Change in Max AL	332.876	185.032	0.177	1.799	.075
Man AL	1	.147	(Constant)	6444.873	1121.702		5.746	.000
			Extraction	−1421.919	1619.845	−0.083	−0.878	.382
Change in Man AL	764.646	198.691	0.364	3.848	.000
Max IMW	1	.087	(Constant)	8155.440	1070.022		7.622	.000
			Extraction	−2849.326	1592.837	−0.166	−1.789	.076
Change in Max IMW	475.383	185.395	0.238	2.564	.012
Man IMW	1	.060	(Constant)	8299.843	1090.147		7.614	.000
			Extraction	−2373.556	1761.559	−0.138	−1.347	.181
Change in Man IMW	594.492	329.411	0.185	1.805	.074

^aAL indicates arch length; IMW, intermolar width; Man, mandible; Max, maxilla; OCP, oral cavity proper; and Std., standard. Significance set at P ≤ .05.

Stepwise multiple regression analysis showed that the combination of mandibular arch length, gender, changes in maxillary intermolar width (IMW), and changes in maxillary arch length significantly influenced the difference in intraoral volume between the two groups, with an adjusted R²of 0.29 being recorded. This indicated that a clinical alteration of those factors influenced the intraoral volume and the impact of having an extraction became less significant.

DISCUSSION

This was the first study to investigate and compare the T0 and T1 volumetric changes of the OCP in patients undergoing extraction and nonextraction orthodontic treatment. The findings indicated that the volume of the OCP increased in all patients, with those undergoing nonextraction treatment experiencing a greater volume increase. The investigation and findings are clinically relevant as potential OCP volume loss in extraction cases may be associated with a potential increased risk of OSA.

Variations in the definition of what constitutes the OCP and the landmarks used to determine its borders made comparisons with other studies challenging.¹⁶ The adoption of the hyoid bone as a landmark, suggested by Halim et al, was not used in the present study as it has been found to be unreliable and potentially negatively influenced by head posture.^16,17 The landmarks selected to create the OCP boundaries in the present investigation were based on previous studies that showed acceptable reliability and reproducibility.¹⁵ Delineation of the oropharynx from the OCP boundaries in the present investigation was in agreement with the anatomical definitions set by Laine and Smoker.¹

The present study indicated that males demonstrated greater changes in OCP volume than females. Previous similar studies did not report on gender and age differences, so comparisons with other investigations was not possible. However, as the mean age of both cohorts was 15 years old, it is reasonable to consider that the male participants were undergoing comparatively greater growth between evaluation timepoints.^18,19 In addition, the morphological differences in craniofacial pattern between male and female participants were likely to influence changes in intraoral volume.^18,19

The changes to mandibular and maxillary arch length resulting from different treatment modalities have been previously reported.^12,20 The change to mandibular arch length was found to be responsible for 15% of the variance of OCP volume between T0 and T1 measurements, whereas the maxillary arch length was responsible for 6% of the variance.

A reduction in mandibular arch length of up to 12.1 mm as a result of premolar extraction has been previously reported.^12,20–23 In the present study, the mandibular arch length in the extraction sample reduced by 0.94 mm. By contrast, the increase in the mandibular arch length in nonextraction patients in the present study of 2.19 mm was comparable with the increase of up to 2.9 mm reported elsewhere.¹² The variation in arch length reduction may have been because the measurement of arch length was from a reliable fixed point (a vertical plane perpendicular to posterior nasal spine) rather than a mobile reference point (such as a plane connecting the contralateral mesial cusps of permanent molars).²⁰

In addition, the combination of the extraction of second permanent premolars and the treatment mechanics associated with the Tip-Edge appliance may have resulted in the minimal arch length reduction observed in the present study. Ongoing growth of the maxilla and mandible may also have negated some of the potential arch length reduction.²⁴

The changes in the mandibular (38%) and maxillary (25%) IMWs were shown to contribute significantly to the differences in the OCP volume in both extraction and nonextraction patients. The reduction of 0.08 mm in the mandibular IMW in the nonextraction group and 2.73 mm in the extraction group were similar to that previously reported.^22,23 The present findings showed a minor reduction in maxillary IMW in the control sample, 0.08 mm, and a more significant reduction in the extraction group of 2.11 mm. This also corresponded to findings in previous studies.^19,22 However, it has been contended that the mesialization of the molars into the relatively narrower anterior dental arch may be responsible for the apparent reduction of the IMW rather than a reduction of the radius of the curve of the arch and a subsequent loss of OCP volume.²⁵ Future studies should consider placing landmarks on all available teeth in the arch or develop a novel way of overcoming the problem of defining the lateral boundaries created in the present study.

A strength of the study was the use of key landmarks that were shown to be reliable and reproducible.

Although the risk of selection bias was identified as a limitation of the study, the risk was minimized by the strict adherence to the inclusion and exclusion criteria. In addition, a lack of reliability in orientation of the image and the selection of the sensitivity threshold are potential risks in all volumetric studies.

A future study is required to further determine the clinical and nonclinical parameters that may influence the OCP volume. Additional research is also required to investigate OCP volume changes in nongrowing patients and in those treated with other fixed and removable orthodontic appliances.

CONCLUSIONS

• The volume of the OCP increased in all patients.

• Nonextraction cases had a greater volume increase.

• Changes to mandibular and maxillary arch length, gender, and maxillary IMW contributed to the changes in the volume of the OCP.