Transverse maxillary deficiency is a common malocclusion frequently encountered by orthodontists, often exhibiting a bilateral or unilateral posterior crossbite.

A common treatment option for this malocclusion is rapid palatal expansion (RPE), which uses various types of jackscrew appliances. In 2006, Park1 introduced the use of orthodontic microimplants to reinforce anchorage during RPE. Microimplant-assisted rapid palatal expansion (MARPE) applies expansion forces directly to the bone, enabling the separation of the midpalatal suture even in adolescents and adults.

External apical root resorption (EARR) is a significant complication of comprehensive orthodontic treatments, including RPE.2 EARR occurs due to external forces applied during orthodontic treatment, resulting in physiological or pathological loss of root dentin or cementum.3 Several factors contributing to EARR have been reported. Most studies have found no significant correlation between age, sex, and root resorption.4-18 Various results have reported the relationship between tooth length and root resorption; however, studies revealed that EARR increases with shorter teeth19-21 while others have indicated that resorption increases with longer teeth.4,5,22-24 Similarly, reports about teeth most vulnerable to EARR are also controversial. Consequently, while some have reported that the central incisors are at greater risk,25-28 others have reported that the lateral incisors are more susceptible to root resorption.6,22,28-31

Most studies analyzing the relationship between root resorption and palatal expansion have focused primarily on the anchorage tooth. Earlier research observed root resorption only in the first premolar and first molar, which serve as the anchorage of RPE.32,33 However, recent studies using cone-beam computed tomography (CBCT) reported that root resorption also occurs in the second premolars that were excluded from the anchorage unit.34,35

Few researchers have reported the effects of palatal expansion on the anterior teeth. Sameshima and Sinclair6 found that palatal expansion did not affect root resorption of the maxillary anterior teeth, regardless of rapid or slow expansion. Lam et al.36 assessed the prognosis of maxillary anterior teeth following palatal expansion by measuring pulp blood flow (PBF) and concluded that although a decrease in anterior PBF was observed post-expansion, it was not statistically significant. Vardimon et al.37 evaluated anterior teeth root resorption caused by palatal expansion using fluorescent microscopy and found significant resorption only in central incisors.

In summary, there is a lack of comprehensive research on the effects of palatal expansion on the non-anchored anterior teeth. Therefore, the purpose of this study is to measure and compare EARR in the maxillary central and lateral incisors following expansion with MARPE using CBCT and to evaluate correlations between sex, age, amount of expansion, initial tooth length, root/crown ratio, and root resorption.

This clinical study was approved by the Institutional Review Board of the Dankook University Dental Hospital (DKUDH IRB 2024-04-001).

Participants

The study included patients who had undergone full skull CBCT scans before and after palatal expansion at Dankook University Dental Hospital between April 2015 and August 2023. In total, 60 patients (22 males and 38 females) met the inclusion criteria (Table 1).

Table 1 . Participant characteristics.

	Mean age	Standard deviation
Male (n = 22)	18.05	4.98
Female (n = 38)	15.89	5.28

CBCT scans were taken before expansion (T1) and 3 to 6 months after expansion (T2). The amount of expansion was measured as the width difference between the center of resistance of the upper and lower first molars on the CBCT images taken before expansion.

The inclusion criteria for the study were as follows: (1) Patients who had undergone full skull CBCT scans before and after expansion, and (2) CBCT images of high quality that allowed for clear identification of the maxillary incisors root.

The exclusion criteria were as follows: (1) Patients with fixed appliances attached to the maxillary incisors during the retention period following expansion, (2) a history of trauma to the maxillary incisors, (3) impacted maxillary incisors, (4) severe craniofacial deformities such as cleft lip and palate, and (5) a history of previous orthodontic treatment or orthognathic surgery.

CBCT analysis

All CBCT images were taken using the Alphard VEGA (Asahi Roentgen Ind. Co. Ltd, Kyoto, Japan) system under the following conditions: tube voltage (80 kVp), tube current (6.0 mA), voxel size (0.39 mm), scan time (17 seconds), and a field of view measuring 200 mm in height and 179 mm in depth. The CBCT data were exported in Digital Imaging and Communication in Medicine (DICOM) format and analyzed using InVivo Dental 5 software (version 5.3; Anatomage, San Jose, CA, USA).

Tooth length, root/crown ratio, resorption length, and resorption length percentage

CBCT images were re-oriented for measurements using the following steps (Figure 1): (1) In the axial view, the reticle was aligned to pass through the widest part of the incisor’s buccolingual and mesiodistal regions. (2) In the coronal view, the reticle was positioned to pass through both the apex and the midpoint of the incisal edge. (3) In the sagittal view, actual measurements were taken. In this view, a cementoenamel junction (CEJ) line was set, connecting the buccal and palatal CEJ points. The axis line was defined as the line connecting the apex to the midpoint of the incisal edge, with the intersection of these two lines defined as point X.

Figure 1. Re-orientation. A, Axial view: the reticle is set to pass through the widest part of the incisor’s buccolingual and mesiodistal regions. B, Coronal view: the reticle is set to pass through the apex and midpoint of incisal edge. C, Sagittal view: measurements were performed.
L, tooth length; r, root length; c, crown length.

Tooth length was defined as the distance from the midpoint of the incisal edge to the apex of the incisor. Root length (r) was measured as the distance from point X to the apex of the incisor, while crown length (c) was the distance from point X to the midpoint of the incisal edge (Figure 1). The root crown ratio (R), measured only at T1, was calculated as follows:

Resorption length (RL) was defined as the difference in tooth length (L) between T1 and T2. The resorption length percentage (RLp) was calculated using the following formula:

Tooth volume, resorption volume, resorption volume percentage

To measure tooth volume, each tooth was reconstructed using the InVivo Dental 5 software (Figure 2). The program’s volume measurement function was utilized, with the Hounsfield unit (HU) value set between 788.5 and 3,000. HU is a relative measurement based on air (–1,000) and water (0). Given that the typical HU range for trabecular bone adjacent to the root is between 300 and 800, the HU value for teeth is expected to be higher. Therefore, the program’s default minimum value of 788.5 was considered appropriate for measurement, while 3,000 is the maximum HU value allowed by the program.

Figure 2. Measurement of center of resistance (Cre) by connecting the first molars’ bifurcation point before (A) and after (B) microimplant-assisted rapid palatal expansion. The Frankfort Horizontal plane was used as the horizontal reference plane. The amount of expansion was defined as the difference in Cre measured between T1 and T2 period.

Resorption volume (RV) was defined as the difference in tooth volume (V) between T1 and T2. The resorption volume percentage (RVp) was calculated as follows:

Center of resistance, amount of expansion

To calculate the amount of expansion, a modified version of the Yonsei Transverse Index was used. The center of resistance (Cre) was calculated as the distance from both maxillary first molars’ center of resistance (bifurcation point) (Figure 3). The Frankfort Horizontal plane served as the horizontal reference plane. The amount of expansion (E) was defined as the difference in Cre measurements between T1 and T2.

Figure 3. Measurement of tooth volume (V) by InVivo Dental 5 software.

Statistical analysis

A minimum of 11 patients were required to conduct a paired sample t test with a minimum power of 80%, a significance level of 0.05, and an effect size of 0.85. The sample size calculation was performed using G*power 3.1.9.4 for Windows (Heinrich-Heine-University, Düsseldorf, Germany). Ultimately, 60 patients, 120 central incisors, and 120 lateral incisors were included in the study.

All measurements were conducted by one examiner (JHC). Statistical analysis was performed using SPSS software for Windows, version 25.0 (IBM Corp., Armonk, NY, USA), with the significance level set at 5% (P < 0.05) for all analyses.

To evaluate intra-examiner reliability, 10 samples were randomly selected for re-measurement two weeks after the initial measurements. Intraclass correlation coefficients (ICCs) were calculated. Furthermore, descriptive statistics were used to measure means and standard deviations for all analyses. The Shapiro–Wilk test was employed to test normality, and Levene’s test was used to evaluate data homogeneity. A paired sample t test was conducted to compare incisor length and volume before and after expansion. An independent sample t test was performed to examine differences based on sex between central and lateral incisors and between the left and right sides. Pearson correlation analysis was used to evaluate the relationships between variables such as age, amount of expansion, pre-expansion tooth length, root/crown ratio and RL, RLp, RV, RVp.

The intra-examiner reliability values, measured using ICC with 95% confidence intervals, were above 0.924 for all measurements, indicating high reliability.

Descriptive analyses of tooth length, root length, and crown length of the central and lateral incisor are presented in Table 2.

Table 2 . Descriptive analysis of tooth length, root length and crown length of central and lateral incisor.

Variable		Mean ± standard deviation	Minimum	Maximum
Central incisor (n = 120)	L	22.91 ± 1.67	17.67	27.79
	r	11.31 ± 1.59	7.24	16.51
	c	11.60 ± 0.84	9.59	13.78
Lateral incisor (n = 120)	L	22.06 ± 1.86	18.11	26.17
	r	12.17 ± 1.51	8.37	16.07
	c	9.89 ± 1.10	6.30	12.25

L, tooth length; r, root length; c, crown length..

An independent t test revealed no significant differences between the left and right measurements (P > 0.05). Consequently, right and left central and lateral incisor sample were pooled into central and lateral incisor samples, respectively (Table 3).

Table 3 . Independent t test results according to the right and left side.

		Mean	SD	P value
L (T1, central incisor)	Rt	22.88	1.66	0.864
	Lt	22.93	1.69	0.864
R (T1, central incisor)	Rt	0.98	0.16	0.935
	Lt	0.98	0.17	0.935
V (T1, central incisor)	Rt	786.28	128.02	0.863
	Lt	782.21	129.11	0.863
L (T2, central incisor)	Rt	22.31	1.62	0.874
	Lt	22.36	1.63	0.874
V (T2, central incisor)	Rt	737.20	120.92	0.975
	Lt	737.90	119.81	0.975
L (T1, lateral incisor)	Rt	22.08	1.81	0.895
	Lt	22.04	1.92	0.895
R (T1, lateral incisor)	Rt	1.25	0.21	0.924
	Lt	1.24	0.22	0.924
V (T1, lateral incisor)	Rt	554.37	105.11	0.548
	Lt	542.02	118.95	0.548
L (T2, lateral incisor)	Rt	21.72	1.78	0.628
	Lt	21.56	1.89	0.628
V (T2, lateral incisor)	Rt	537.65	101.35	0.584
	Lt	526.81	114.47	0.584

L, tooth length; R, root/crown ratio; V, tooth volume; SD, standard deviation..

The paired t test showed a significant difference in the tooth length and volume of the central and lateral incisors before and after expansion, indicating that root resorption occurred after expansion (Table 4).

Table 4 . Comparison by paired t test for tooth length and volume, before and after expansion and resorption, and comparison by independent t test for resorption index (RL, RLp, RV, RVp) between central and lateral incisor.

Teeth	n	Pre-expansion (T1)			Post-expansion (T2)		P value†	Difference (T1–T2)		P value‡
		Mean	SD		Mean	SD		Mean	SD
Central incisor (L)	120	22.97	1.56		22.39	1.49	***	0.57 (RL)	0.58	0.026
								2.47 (RLp)	2.44	0.046
Central incisor (V)	120	784.25	128.04		737.55	119.86	***	46.70 (RV)	38.56	0.000
								5.82 (RVp)	4.56	0.000
Lateral incisor (L)	120	22.06	1.86		21.64	1.83	***	0.42 (RL)	0.46	0.026
								1.88 (RLp)	2.03	0.046
Lateral incisor (V)	120	548.19	111.94		532.23	107.79	***	15.96 (RV)	19.92	0.000
								2.81 (RVp)	3.31	0.000

L, tooth length; V, tooth volume; RL, resorption length; RLp, resorption length percentage; RV, resorption volume; RVp, resorption volume percentage; SD, standard deviation..

***P < 0.001..

^†Paired t test was performed..

^‡Independent t test was performed..

Furthermore, an independent t test showed no significant sex difference in the values of RL, RLp, RV, and RVp for the central and lateral incisors.

Independent sample t test also revealed that RL, RLp, RV, and RVp were significantly higher in the central incisor compared with the lateral incisor, suggesting that the central incisor shows more root resorption than the lateral incisor after expansion (Table 4).

Pearson correlation analysis revealed a significant correlation between age and RLp of the central incisors (P = 0.048) (Table 5). The scatterplot shows a positive coefficient, indicating that as age increases, the RLp of the central incisors also increases (Figure 4).

Figure 4. Pearson correlation scatterplot. The scatterplot shows a positive coefficient, which indicates that as age increases, the resorption length percentage of the central incisors (RLp1) increases.

Table 5 . Correlation between age, amount of expansion and central incisor’s resorption index (RLp, RVp).

Central incisor		RLp	RVp
Age	Pearson correlation	0.181*	−0.133
	Sig. (2-tailed)	0.048	0.148
E	Pearson correlation	−0.086	0.225*
	Sig. (2-tailed)	0.351	0.014

E, amount of expansion; RLp, resorption length percentage; RVp, resorption volume percentage..

Correlation significance: *P < 0.05..

Furthermore, there was a significant correlation between the amount of expansion and the RVp of the central incisors (P = 0.014) (Table 5). The corresponding scatterplot also shows a positive coefficient, indicating that as the amount of expansion increases, the RVp of the central incisors increases accordingly (Figure 5).

Figure 5. Pearson correlation scatterplot. The scatterplot shows a positive coefficient, which indicates that as the amount of expansion (E) increases, the resorption volume percentage of the central incisors (RVp1) increases.

Moreover, a significant correlation was found between the pre-expansion tooth length of the central incisors and both the RL and RLp of the central incisors (P = 0.001/P = 0.007) (Table 6). The scatterplot indicates positive coefficients, indicating that longer initial central incisor tooth lengths are associated with greater RL and RLp of the central incisors (Figure 6).

Figure 6. Pearson correlation scatterplot. The scatterplot shows a positive coefficient, which indicates that the longer the initial central incisor tooth length (L1), the greater the resorption length of the central incisor (RL1) and resorption length percentage of the central incisors (RLp1).

Table 6 . Correlation between central incisor’s initial tooth length, root/crown ratio, and resorption index (RL, RLp).

Central incisor		RL	RLp
L	Pearson correlation	0.307**	0.247**
	Sig. (2-tailed)	0.001	0.007
R	Pearson correlation	0.190*	0.157
	Sig. (2-tailed)	0.038	0.087

L, tooth length; R, root/crown ratio; RL, resorption length; RLp, resorption length percentage..

Correlation significance: *P < 0.05, **P < 0.01..

Similarly, a significant correlation was observed between the root/crown ratio of the central incisors before expansion and the RL (P = 0.038) (Table 6). The scatterplot shows a positive coefficient, indicating that a higher initial root/crown ratio corresponds to a greater RL in the central incisor (Figure 7).

Figure 7. Pearson correlation scatterplot. Scatterplot shows a positive coefficient, which indicates that the higher the root/crown ratio of the initial central incisor (R1), the greater the resorption length of the central incisor (RL1).

For the lateral incisors, a significant correlation was noted between pre-expansion tooth length and RL (P = 0.041) (Table 7). The scatterplot shows a positive coefficient, indicating that longer initial lateral incisor tooth lengths are associated with greater RLs (Figure 8).

Figure 8. Pearson correlation scatterplot. Scatterplot shows a positive coefficient, which indicates that the longer the initial lateral incisor tooth length (L2), the greater the resorption length of the lateral incisors (RL2).

Table 7 . Correlation between lateral incisor’s initial tooth length and resorption index (RL).

Lateral incisor		RL
L	Pearson correlation	0.187*
	Sig. (2-tailed)	0.041

L, tooth length; RL, resorption length..

Correlation significance: *P < 0.05..

In contrast, the root/crown ratio of the lateral incisors before expansion did not show a significant correlation with other root resorption values.

Previous studies on root resorption associated with palatal expansion were limited to the anchorage teeth. Langford and Sims32 used the first premolar and first molar as anchorage units in RPE. After extracting the first premolar immediately after expansion, scanning electron microscope (SEM) analysis revealed root resorption on the buccal surface of the root. Similarly, Barber and Sims33 analyzed the first premolars extracted immediately after expansion, using SEM, reporting that root resorption occurred only when the premolar was used as an anchorage unit.

Baysal et al.34 investigated whether root resorption occurs not only in anchorage teeth but also in non-anchorage teeth. Root resorption was assessed by comparing the tooth volume before and immediately after expansion using CBCT. The result showed that in addition to the first premolar and first molar, root resorption was also observed in the second premolar, a non-anchorage tooth. Leonardi et al.35 further compared the root length and volume using CBCT before expansion and six months after expansion. They also confirmed that root resorption occurred in the non-anchorage second premolar and subsequently noted root resorption in the second premolar when bone-borne RPE was used.

Few studies specifically examined the effects of palatal expansion on anterior teeth. Sameshima and Sinclair5 evaluated anterior tooth root resorption following palatal expansion using periapical radiographs. They reported that palatal expansion did not impact root resorption of the maxillary anterior teeth, regardless of rapid or slow expansion protocol use. However, their use of two-dimensional periapical radiographs for measuring root resorption may have posed accuracy limitations. In a study by Lam et al.,36 PBF in the maxillary anterior teeth following palatal expansion with RPE or MARPE was evaluated. PBF was measured pre-expansion, two weeks post-expansion, and three months after expansion using laser Doppler flowmetry.

The result showed a significant reduction in PBF in the anterior teeth two weeks after expansion in both RPE and MARPE groups. The PBF value recovered after three months of expansion; however, it remained lower than pre-expansion levels, which was not statistically significant. Vardimon et al.37 conducted palatal expansion on 14 cats to evaluate anterior tooth root resorption using fluorescent microscopy 120 days after completion of the expansion. Their results revealed significant root resorption only in the central incisors. The authors attributed this to root proximity, which resulted from mesial crown tipping of the central incisors. A significant correlation was found between tip angle, root proximity, and root resorption of the central incisor.

In contrast, root resorption in the lateral incisors was not significant when compared to the control group. The authors noted the challenges of accurately evaluating root resorption using two-dimensional radiographs.

In this study, using CBCT for measurement, significant root resorption was observed in the maxillary incisors following expansion. Most previous studies assessed root resorption during comprehensive orthodontic treatment using periapical radiographs. For instance, Chung et al.4 and Baumrind et al.7 reported root resorption of the central incisors measuring 1.63 and 1.36 mm, respectively. Nanekrungsan et al.29 and Mirabella and Artun24 found central incisor resorption to be 1.39 and 1.47 mm, while lateral incisors experienced resorption of 1.69 and 1.63 mm, respectively. Similarly, Jose et al.38 used CBCT to evaluate root resorption of maxillary incisors during en masse retraction with temporary anchorage devices following maxillary first premolar extractions, reporting resorption of central and lateral incisors to be 1.397 and 1.330 mm, respectively. Considering these findings, we can conclude that approximately one-third of the total root resorption of maxillary incisors, which typically occurs during comprehensive orthodontic treatment, may take place during the process of spontaneous space closure following palatal expansion.

Regarding tooth sensitivity to root resorption, previous studies have reported maxillary central incisors or lateral incisors as the most vulnerable. Several studies6,22,28-31 reported that root resorption occurs most frequently in maxillary lateral incisors, and due to the high incidence rate of root abnormalities such as dilaceration, microdontia, dens invaginatus, lateral incisors are vulnerable to root resorption.29 In contrast, Jung and Cho25 found that the average reduction in root length of maxillary central incisors was greater than that of other teeth, explaining that central incisors generally move greater distances during treatment, increasing the likelihood of root resorption. Furthermore, Vardimon et al.37 observed significant root resorption only in central incisors after expansion.

Our research, however, reveals that significant root resorption was observed not only in the central incisors but also in lateral incisors, which contrasts with some previous findings. In regards to the effect of sex on root resorption, although some studies have reported increased resorption in male7 or female,39 most studies have shown that sex does not significantly affect root resorption.4-6,14-17

Our study confirms that sex had no significant effect on the maxillary incisor's root resorption following palatal expansion.

The majority of previous studies report no relationship between age and root resorption during orthodontic treatment.4-13 In contrast, some studies claim that resorption increases with age.14,15

Our research also confirms that as age increases, root resorption (RLp) of central incisors increases. According to Schour and Massler,18 maxillary central and lateral incisor root development is completed by the age of 10 to 11 years. If root resorption occurs before this period, recovery may be expected through additional growth in root length. However, after root development is complete, resorption ultimately results in a decrease in root length.

The amount of expansion had a positive correlation with root resorption (RVp) of the central incisors, but not the lateral incisors. Vardimon et al.37 suggested that most of the space created by expansion occurs between the central incisors; the crown mesial tipping movement and subsequent root resorption are relatively bigger than lateral incisors.

The relationship between tooth length and root resorption has been reported in various ways. Some studies have suggested that resorption increases with shorter teeth.19-21 Picanço et al.19 hypothesized that shorter root lengths and the lower root/crown ratio result in greater forces being applied to the root apex by the relatively larger crown. Recently, several studies have reported that resorption increases with longer teeth.4,5,22-24 The authors explain that the longer the root length, the greater the force is required when moving the tooth, and from the same amount of force, relatively greater torque is applied to the apex of the tooth. Also, from the same torque, more root apex displacement occurs.

In this study, root resorption increased as the initial tooth length of both the central and lateral incisors increased. Vardimon et al.37 explained that the cause of the incisor's root resorption after palatal expansion was caused by compression forces due to the crown mesial tipping and root distal tipping movement. Generally, the center of resistance of an incisor is located 1/4 to 1/3 apically from the CEJ, and as root resorption increases, the center of resistance moves coronally.40 Considering that the transseptal fibers, which facilitate spontaneous space closure of diastema, are located across the cementum between adjacent teeth near the CEJ,41 In normal central incisors, the forces exerted by the transseptal fibers are applied coronal to the center of resistance, causing crown mesial tipping and root distal tipping movement. However, when root resorption occurs, the center of resistance moves to the coronal side. Consequently, transseptal fiber forces are applied relatively more apically to the center of resistance, thereby reducing crown mesial tipping and root distal tipping movement (Figure 9).

Figure 9. Comparison of tipping movement and center of resistance (red dots) change between before and after root resorption. A, In normal central incisors, the force of the transseptal fibers (red lines) is applied relatively coronal to the center of resistance, causing crown mesial tipping and root distal tipping movement. B, When root resorption occurs, the center of resistance moves to the coronal side. Consequently, the force of the transseptal fibers is applied relatively more apically to the center of resistance, reducing crown mesial tipping and root distal tipping movement.

In summary, the shorter the initial root length of the tooth, the smaller the crown mesial tipping and root distal tipping movement is induced by the transseptal fibers, which might result in reduced compression force on tooth apex and surrounding alveolar bone area.

A comparison of maxillary incisors using CBCT before and after MARPE revealed the following:

First, significant maxillary incisor root resorption was observed after MARPE, and root resorption was significantly higher in central incisors than the lateral incisors.

Second, several factors were identified as risk factors for EARR after MARPE, including initial age, a larger amount of expansion, initial tooth length, and root/crown ratio. Based on these results, we can infer that patients with short maxillary incisor roots are not contraindicated for MARPE.

Conceptualization: JHC. Data curation: JHC. Formal analysis: All authors. Investigation: JHC. Methodology: All authors. Project administration: All authors. Resources: All authors. Supervision: All authors. Validation: JHC. Visualization: JHC. Writing–original draft: JHC. Writing–review & editing: All authors.

No potential conflict of interest relevant to this article was reported.

None to declare.