Variability associated with maxillary infrazygomatic crest and palatal bone width, height, and angulation in subjects with different vertical facial growth types: a retrospective cone-beam computed tomography study
To assess the infrazygomatic crest (IZC) and palatal bone width, height, and angulation in patients with different vertical facial growth types as potential miniscrew insertion sites. In this retrospective cone-beam computed tomography study, 162 subjects (81 males and 81 females, mean age 16.05 ± 0.65 years) were included. They were divided into three groups (hypodivergent, normodivergent, and hyperdivergent) based on the Frankfort mandibular plane angle. Ten buccal bone measurements were made at two different coronal sections: maxillary first molar mesiobuccal and distobuccal roots (bilaterally). Six palatal bone measurements were made on a sagittal section at the maxillary central incisors (bilaterally). A total of 32 measurements per subject were considered in the study. No significant difference was observed for the IZC (width and angle) at the maxillary first molar mesiobuccal root. A comparison of normodivergent and hyperdivergent groups for buccal width at the distobuccal root of the first molar showed significant differences. Palatal bone thickness at the level of 2 mm distal to the apex of the central incisor was significantly higher for the hyperdivergent group (10.43 mm) compared with the normodivergent (7.58 mm) and hypodivergent groups (7.83 mm). Hyperdivergent subjects tend to present a longer and deeper IZC and increased palatal bone thickness compared with other groups. The recommended insertion angle for the IZC mini-implant at 3 mm from the alveolar crest should be between 75.5° and 77°.ABSTRACT
Objectives
Materials and Methods
Results
Conclusions
INTRODUCTION
Anchorage control is the key to successful orthodontic treatment. The use of miniscrews to obtain absolute anchorage has been demonstrated in clinical orthodontics and has had promising results. During the past decade, miniscrews have been successfully used to obtain absolute anchorage in various malocclusions. However, a large body of evidence shows a miniscrew failure rate between 14% and 20% and directly correlated with insertion site anatomy.1 In addition, it has been shown that miniscrews in the palate have a higher success rate than do interradicular miniscrews in the maxilla and mandible.2–4 When considering ease of placement and application of direct orthodontic force, maxillary and mandibular interradicular sites are still the preferred locations for miniscrews. However, interradicular miniscrews in the mandibular posterior zone have the highest failure rate, about 20% to 29%.5–8 In addition, interradicular placement may damage the adjacent tooth roots or interfere with the desired orthodontic tooth movement.
According to De Clerck et al.,9 the infrazygomatic crest (IZC), the inferior border of the maxillary zygomatic buttress, is the preferred site for the placement of miniscrews due to its location, solid bone structure, and safe distance from maxillary molar roots. Anatomically, the IZC has two cortical plates: a buccal plate and the lateral wall of the maxillary sinus. This anatomical advantage allows bicortical fixation and improves the primary stability of the miniscrew.10,11 However, the IZC area is only 2- to 5-mm thick, and using a standard 5- to 7-mm length miniscrew may cause perforation of the maxillary sinus during placement.12
Various studies have measured palatal bone thickness and indicated that the anterior paramedian palate provides sufficient vertical bone height for the safe and secure placement of miniscrews.13,14 These infrazygomatic and palatal bone thickness measurements vary in subjects of different races, growth patterns, gender, and growth status. Due to the frequent use of skeletal anchorage, it is extremely important to conduct studies to assess the thickness of the IZC and palate to better understand anatomical dimensions, providing safer surgical procedures and minimizing possible failures.
Therefore, this study aimed to analyze infrazygomatic crest and palatal bone width, height, and angulation in different vertical facial growth types to aid clinicians in planning and for successful placement of miniscrews.
MATERIALS AND METHODS
This retrospective study was approved by the University of Louisville Institutional Review Board. The study sample included pretreatment cone-beam computed tomography (CBCT) scans taken at three private practice orthodontists in Louisville, Kentucky, USA, from 2011 to 2021. CBCTs were acquired identically, using an i-CAT scanner with 110 kVp, 60 mA, 0.3-mm voxel size, a scan time of 18 seconds, and a field of view of 17 × 23 cm.
The inclusion criteria were normal craniofacial development, full permanent dentition except for third molars, and absence of previous orthodontic or orthognathic surgical treatment. Patients who presented one of the following criteria were excluded from the study: distorting pathology affecting the maxilla, periodontal or endodontic disease affecting the bone, and the presence of metallic artifacts in the maxillary region.
Following the inclusion and exclusion criteria, 27 patients were chosen randomly from both males and females of each facial growth pattern (162 patients in total) based on Frankfurt mandibular plane angle (FMPA): hyperdivergent (HrD; FMPA ≥28°), normodivergent (NoD; FMPA = 22–28°), and hypodivergent (HoD; FMPA ≤22°). Detailed demographic data are provided in Table 1. A sample size of n = 54 per group was deemed sufficient to detect a Cohen’s effect size of at least 0.6 with 80% power at an alpha level of 0.05. This effect size was consistent with results from previous studies.15,16

CBCT measurements were taken using Invivo 6 (Anatomage, Inc., Santa Clara, Calif). Measurement sites were developed using methods previously described in the literature.10,17 Ten buccal bone measurements were taken at a coronal slice oriented through the maxillary first molar mesiobuccal root (M1–M5) and distobuccal root (D1–D5), bilaterally (Figure 1). Six palatal bone measurements were taken at a sagittal section oriented through the maxillary central incisor (P1–P6), bilaterally (Figure 2). Overall, a total of 32 measurements (20 for IZC and 12 for palatal bone) were recorded for each patient. Figure 3 shows CBCT screenshots of molar and palatal measurements of a test patient.



Citation: The Angle Orthodontist 94, 3; 10.2319/062023-430.1



Citation: The Angle Orthodontist 94, 3; 10.2319/062023-430.1



Citation: The Angle Orthodontist 94, 3; 10.2319/062023-430.1
Each CBCT was measured by one of three coinvestigators, and each investigator also duplicated measurements of 10 patients measured by another to assess interrater reliability.
Statistical Analysis
Measurements were first tested for significant differences across the midline. Using a paired-sample t-test, each of the 16 right and left measurement pairs within each group (HrD, NoD, HoD) were compared. The null hypothesis was that there would be no significant differences between right- and left-side measurements in any group. Descriptive statistical analysis was performed for buccal and palatal bone measurements based on the growth pattern and gender and reported mean and standard deviatioin in Table 2. A one-way analysis of variance (ANOVA) was used to compare means across these three groups. Tukey’s honestly significant difference (HSD) test was applied for post hoc pairwise comparisons. Finally, the interclass correlation coefficient (ICC) was calculated for the interrater reliability. A P value <0.05 was deemed to be a statistically significant result. All analyses were performed in SPSS (version 28; IBM, Armonk, NY).

RESULTS
Interrater reliability was good to excellent, with the ICC ranging between 0.774 and 0.998 for all three investigators. Comparing the right and left sides, only two parameters (M2 and P1) from HrD and one parameter (P6) from NoD were statistically significant, with mean differences of 0.41 mm, 0.76 mm, and 0.47 mm, respectively. Due to the small mean differences and no pattern, these were determined to be similar enough to combine. Therefore, each right and left measurement pair was averaged together to create a single (average) measurement.
Infrazygomatic Crest Measurements
Buccal bone measurements at the mesiobuccal root of the first molar showed nonsignificant (P > .05) results on comparison among the HrD, NoD, and HoD groups. Most of the first molar distobuccal root also showed nonsignificant results on intergroup comparisons. Only two of the distobuccal root measurements, D3 (P = .027) and D4 (P = .013), were statistically significant on comparison of NoD and HrD. The mean difference of the D3 measurements between NoD and HrD was 0.71 mm, with means of 7.61 mm and 8.31 mm, respectively. In contrast, the mean difference of the D4 measurements between NoD and HrD was 0.64 mm, with means of 7.67 mm and 8.20 mm, respectively. For both D3 and D4, the HoD population was not significantly different from the other two groups, and the NoD group was smaller than the HrD group. It should be noted that D3 and D4 were geometrically very similar measurements and often coincided (Figure 4, Table 2).



Citation: The Angle Orthodontist 94, 3; 10.2319/062023-430.1
No significant difference was noticed for the angular measurements M5 and D5 on the multiple comparisons of all three groups. The mean angle of insertion at the mesiobuccal root of the first molar was 77.0° whereas it was 75.5° at the distobuccal root of the first molar.
Palatal Measurements
Of the six palatal measurements, three showed significant differences: P2 (P < .001), P3 (P = .003), and P6 (P = .016). P2 was significantly different between HrD and HoD (mean difference = 2.6 mm) and HrD and NoD (mean difference = 2.84 mm). No significant difference was observed between HoD and NoD (P > .05). The mean values of P2 for HrD, HoD, and NoD were 10.43 mm, 7.83 mm, and 7.58 mm, respectively.
P3 was significantly different between HrD and NoD (mean difference = 1.50 mm). The mean values for the HrD and NoD groups were 10.96 mm and 12.47 mm, respectively. The HoD group was not significantly different from the other two groups, with a mean value of 11.56 mm.
P6 was significantly different between the HrD and HoD groups (mean difference = 0.67 mm). The mean values of HrD and HoD were 3.78 mm and 3.11 mm, respectively. The NoD group was not significantly different from the other two groups, with a mean value of 3.35 mm (Figure 4).
DISCUSSION
The use of miniscrews has been proven to be an effective method of skeletal anchorage when moving teeth in orthodontic treatment. However, they have been shown to have a failure rate of 13.5%.1 Factors influencing the success rate of miniscrews include but are not limited to root proximity, sinus perforation, insertion site, type of bone, and screw length. Choi et al.18 indicated that failure rates tended to be higher in the mandible than the maxilla due to excessive insertion torque, increasing the chances of necrosis of surrounding bone. Thus, the placement of miniscrews is often viewed as technique sensitive, discouraging practitioners from using them.
Two placement sites that have shown high success rates are the palate and IZC.13,19 However, Jia et al. indicated that the length of miniscrews in the IZC is correlated to higher failure due to sinus perforation.20 The current study is the first to analyze bone dimensions in both the paramedian palate and IZC in groups with different vertical growth patterns to optimize miniscrew placement.
This study showed no statistically significant differences among the three groups for any of the measurements taken at the first molar mesial root. Oksayan et al.21 showed that high-angle subjects had lower sinus volumes than low-angle patients did, indicating that there was an increase in alveolar bone height. Increased bone height would decrease the risk of sinus perforation from miniscrew placement. In contrast, studies performed by Murugesan and Jain16 found that IZC bone thickness was reduced in high-angle patients, with statistically significant differences between high and low-angle patients above the first molar. Such differences could be attributed to the sample population and study methodology.
Distobuccal root measurements D3 and D4 showed significant differences between HoD and NoD samples. The HoD sample showed higher mean values than the NoD sample did. This may be due to stronger masticatory force in low-angle patients that stimulates bone remodeling in this area. According to Kiliardis et al.,22 bone structures surrounding areas of muscular activity tend to adapt by bone deposition, particularly when tension exceeds a threshold. Although not statistically significant, HrD subjects also had higher mean D3 and D4 measurements than NoD subjects did. Husseini et al.23 found similar data and ascribed this difference to HrD subjects experiencing more vertical growth and subsequent eruption of molars, increasing IZC height instead of buccopalatal width. In addition, previous studies demonstrated that more distal locations in the IZC region show gradually increasing bone thickness, which would magnify the discrepancies that were seen among groups in the current study.24,25
The palatal measurement P2 showed significantly greater mean values in the HrD groups than in both NoD and HoD groups. It is possible that central incisor proclination could increase the P2 distance by tipping the root more posteriorly and changing reference point Y (Figure 2), resulting in the higher value seen in HrD subjects. Palatal bone thickness at P3 was significantly lower in the HrD population than in NoD. Ozdemir et al.15 found that the cortical bone thickness of the palate in low-angle subjects was significantly higher than in high-angle subjects. Wang et al.16 found that the anterior portion of the paramedian palate had sufficient bone volume for miniscrews in all three vertical growth types. However, high-angle patients had significantly less bone in the posterior areas. This suggests that the growth pattern along with the site of mini-screw insertion greatly affects the success rate.
Lastly, the palatal bone thickness at the first molar (P6) was statistically significantly different between HrD and HoD, with the mean HrD higher. However, these results were clinically nonsignificant. One explanation for the statistical difference could be the same as mentioned earlier, in which compensatory eruption increases alveolar bone height. However, Suteerapongpun et al.26 and Johari et al.27 found that open-bite skeletal patterns had a statistically significant decrease in bone thickness compared with normal growth patterns at the level of the first molar palatal root. Differences in these results could be attributed to variations in the methods and inclusion criteria.
The findings of this study will allow practitioners to understand the effect of facial type on the IZC and palatal bone dimensions for use as potential miniscrew insertion sites. This study also reflected the recommended angle of insertion for the IZC to avoid anatomical landmarks such as roots and to ensure maximum bone volume for the stability of miniscrews. Although a radiologic examination is always recommended before extra alveolar miniscrew insertion, the results of this study demonstrated that IZC and palatal miniscrews need to be longer for HrD patients and medium length for NoD and HoD patients.
CONCLUSIONS
-
Hyperdivergent subjects tend to present a longer and deeper IZC and increased palatal bone thickness compared with the other groups.
-
The recommended angle of insertion for the IZC is between 75.5° and 77° to avoid damage to the buccal roots of the maxillary first molar and to have maximum bone thickness for the stability of infrazygomatic mini-implants.
-
Palatal bone thickness at the level of 2 mm distal to the apex of the central incisor was statistically higher for the hyperdivergent group (10.43 mm) compared with the normodivergent (7.58 mm) and hypodivergent groups (7.83 mm).

Diagram depicting measurements taken of alveolar bone taken at molar roots. First, three references were identified: occlusal plane, the line bisecting the MB cusps of both maxillary first molars; point X, the lateral-most point of buccal bone horizontal to the mid root point of the cross-sectioned root; and point S, the intersection of sinus floor with a line tangent to the molar buccal root surface. (A) M1 is the horizontal distance between the root and buccal plate at midroot level (intersecting point X). M2 is the horizontal distance between the root and buccal plate, taken at the apex level. M3 is the smallest distance between point X and the sinus floor, staying 1.0 mm away from the root apex. (B) M4 is the distance between point X and point S. (C) M5 is the angular measurement of the intersection between M3 (extended) and the occlusal plane. As noted previously, points D1–D5 correspond with measurements M1–M5 but taken at the distobuccal root.

Diagram depicting measurements taken of the paramedian palatal bone. (A) P1 is the distance between the superior and inferior borders of the maxilla, measured 2.0 mm away from the central incisor apex and perpendicular to the line intersecting PNS and the palatal alveolar crest. The inferior point of measurement P1 is denoted as point Y. (B) P2 is the distance between point Y and the palatal alveolar crest. P3 is the vertical distance between point Y and the superior border of the maxilla. P4 is the horizontal distance between point Y and a vertical tangent to the facial surface of the incisor. P5 is the vertical thickness of palatal bone at the level of the ipsilateral second premolar. P6 is the vertical thickness of palatal bone at the level of the ipsilateral first molar palatal root. P2–P6 are all taken with the scan oriented in natural head position.

CBCT software screenshots of molar and palatal measurements of an example patient.

Results of linear bone measurements (M1–4, D1–4, P1–6). Standard deviations are shown. All measurements are in millimeters.
Contributor Notes