Proudly Sponsored by The EH Angle Education and Research Foundation

Abstract: The aim of the study was to compare the incidence of sella turcica bridging and sella turcica dimensions in 150 Caucasian subjects who had combined surgical-orthodontic correction of their malocclusion with a randomly selected group of 150 Caucasian subjects who were treated contemporaneously by orthodontic means only. Pretreatment lateral cephalometric radiographs were scanned and analyzed. A sella turcica bridge was identified as a continuous band of bony tissue extending from the anterior cranial fossa to the posterior cranial fossa. The dimensions of the sella turcica were measured. In the group treated by combined surgical-orthodontic means, the incidence of bridging was 16.7%, whereas it was 7.3% in the orthodontics-only group (P = .012). Significant increases in the mean surface area (P = .02) and mean perimeter of the sella turcica (P = .01) were found for the combined surgical-orthodontic group compared with the orthodontics-only group. The mean interclinoid distance was significantly smaller in the surgical-orthodontic group (P = .02). These findings appear to indicate the greater likelihood of sella turcica bridging and abnormal sella turcica dimensions in subjects treated by combined surgical-orthodontic means rather than by orthodontics only.

Key Words: Sella turcica, Bridging.

Accepted: July 2004. Submitted: May 2004

The sella turcica is an important anatomical structure for cephalometric assessment because of its central landmark, sella, a saddle-shaped area of bone located in the middle cranial fossa. The sella turcica lies on the intracranial surface of the body of the sphenoid and consists of a central pituitary fossa bounded anteriorly by the tuberculum sellae and posteriorly by the dorsum sellae. Two anterior and two posterior clinoid processes project over the pituitary fossa. The anterior clinoid processes are formed by the medial and anterior prolongations of the lesser wing of the sphenoid bone, and the posterior clinoid processes represent terminations of the dorsum sellae.

The sella anatomy is variable and has been classified into three types—round, oval, or flat. Variations in the size and shape of the clinoid processes are common. The anterior clinoid processes are larger and more variable. They may be short blunt structures or may project over the pituitary fossa, occasionally uniting. Fusion of the posterior and anterior clinoid processes is known as a sella turcica bridge.

Incidence of bridging has been reported in anatomical and radiographic studies. Direct measurement of the skull¹ and inspection at autopsy² found an incidence of bridging of 5.5% and 6%, respectively, whereas a 4.6% incidence has been reported based on radiographic examination.³ Sella turcica bridging has been classified into two types depending on the type of fusion of the anterior and posterior clinoid processes.⁴ Type A features ribbonlike fusion, and type B is represented by bony extension of the anterior or posterior clinoid process such that they meet or superimpose across the pituitary fossa.

Sella turcica size varies from five to 16 mm in anteroposterior diameter and from four to 16 mm in vertical depth,^5,6 with accepted normal maximum dimensions of 16 mm in anteroposterior dimension and 12 mm in depth.⁷ Normal sella turcica volume has been stated to be 133 mm.^2,7

The incidence of sella turcica bridging in patients with severe craniofacial deviations who required combined surgical-orthodontic treatment has been assessed from lateral cephalometric radiographs.⁴ A sella turcica bridge occurred in 18.6% of 177 subjects, which is more than double the incidence of bridging reported previously in the literature. These data were compared with those of previous studies involving nonorthognathic subjects, where the racial groups may have varied and where assessment of sella turcica bridging was based on anatomical or radiographic means. The latter related to data collected three to four decades previously so that changes in radiographic equipment and technique may have an effect on the differences observed. The anatomical dimensions of the sella turcica may also be abnormal in those who received combined surgical-orthodontic correction of their malocclusion but that was not assessed.

The aim of this study was to compare the incidence of sella turcica bridging and sella turcica dimensions in Caucasian subjects who had undergone combined surgical-orthodontic treatment with a Caucasian group treated contemporaneously by orthodontic means only. The null hypothesis tested was that there was no difference in the incidence of sella turcica bridging or in sella turcica dimensions in Caucasians who had combined surgical-orthodontic treatment compared with Caucasians treated contemporaneously by orthodontic means only.

Pretreatment lateral cephalometric radiographs were retrieved from the case records of 300 Caucasian subjects who were treated at a University Orthodontic Clinic over the previous four years. A total of 150 consecutively treated subjects who had combined surgical-orthodontic treatment were identified from an orthognathic departmental database. The group identified represented all orthognathic subjects treated in this period except those with cleft lip and palate and/ or known craniofacial syndromes, who were excluded because sella turcica dimensions are known to be altered in these patients.^8,9 Although orthognathic surgery may be undertaken for a variety of reasons excluding skeletal deformity, all the subjects in the combined surgical-orthodontic group had treatment to correct a dentofacial deformity. Another 150 subjects, treated contemporaneously by orthodontic means only, were randomly selected from an orthodontic database.

All lateral cephalometric radiographs were taken by trained radiographers in a standardized manner using the same cephalostat (Wehmer Model W-102, Kodak X-Omatic extraoral screen and film) in the University Department of Dental Radiology. Radiographs were of good quality and clearly showed the sella turcica anatomy. A 30 × 60–mm area of the sella turcica region of each radiograph was scanned (Linotype-Hell Office 2 scanner with transparency adaptor, Heidelberg New Color 4000 software, Heidelberg, Germany) at a resolution of 1000 dpi. The radiographs were analyzed using Scion Image for Windows 4.0.2 (Scion Corporation, Frederick, Md). The imaging software was calibrated to account for any differences in magnification due to radiographic technique and/or the scanning process.

Images were assigned an identifier number so that at the time of assessment and analysis, the observers (two experienced clinicians familiar with examining lateral cephalometric radiographs) were unaware of the treatment allocation of each image. Images were randomly selected for analysis in a darkened room. Independently, the assessors identified and classified bridging of the sella turcica according to previously applied criteria.⁴ Type A represents manifest ribbonlike fusion of the anterior and posterior clinoid processes, whereas type B indicates an extension of the anterior and/or the posterior clinoid process, where these two meet anteriorly, posteriorly, or in the middle with a thinner fusion.⁴

The dimensions of the sella turcica were also measured by one observer using the Image analysis software according to the method described by Silverman.⁸ The area (A), perimeter (P), width (W), vertical depth (D), and interclinoid distance (I) were measured. The greatest anteroposterior dimension (GW) was also measured if greater than the width (Figure 1 ).

Cephalometric analysis of those subjects with bridging in each group was undertaken in a darkened room by the same operator who measured sella turcica dimensions. A Numonics Tablet (Numonics Corporation, Montgomeryville, Pa) connected to a personal computer (Dell Dimension XPS T550, Dell Corporation, Austin, Tex) was used.

To assess intra- and interexaminer repeatability in the identification of sella turcica bridging, the 300 lateral cephalometric radiographs were reexamined by each assessor independently one month after the initial analysis. Intraoperator repeatability in measurement of sella turcica dimensions was also assessed one month after initial recordings by retaking measurements of sella turcica dimensions on 75 scanned radiographic images (25% of the total sample). Four weeks after initial cephalometric assessment, the analysis was repeated to allow assessment of intraoperator repeatability.

A chi-square test was used to compare the incidence of sella turcica bridging between the combined surgical-orthodontic group with that of the orthodontics-only group. Two sample t-tests were used to compare mean sella turcica dimensions between the two groups. Two sample t-tests were also used to assess differences between the mean cephalometric values for the combined surgical-orthodontic group with bridging vs the orthodontic group with bridging. The data were checked to confirm that the assumptions necessary for these tests were valid.

Paired t-tests were used to test the mean difference between the original measurements and the repeated measurements for sella turcica dimensions and cephalometric values. The method proposed by Bland and Altman¹⁰ was also used to assess the repeatability. All data were analyzed using SPSS for Windows (Version 11).

There was 100% intra- and interexaminer agreement for assessment of sella turcica bridging on each of the 300 cephalometric radiographs. For each subject, there were no significant differences between the means of the original sella turcica dimensions and the repeated dimensions (P = .081) or between the means of the original and the repeated cephalometric analyses (P = .073). The method of Bland and Altman¹⁰ revealed that the cephalometric values and the sella turcica dimensions were repeatable.

The incidence of bridging in the combined surgical-orthodontic group compared with the orthodontics-only group was 16.7% and 7.3%, respectively (P = .012). In the former group, 40% of bridges were type A and 60% were type B, whereas in the orthodontics-only group, 63.6% were type A and 36.4% were type B.

Details on sella turcica dimensions are shown in Table 1 . The mean sella turcica area in the combined surgical-orthodontic group was 69.1 mm² compared with 65.2 mm² in the orthodontics-only group (P = .02). The mean sella perimeter for each of these groups was 36.7 and 35.5 mm, respectively (P = .01), whereas the mean interclinoid distance in the orthodontics-only group was 4.1 mm compared with 3.5 mm for the surgical-orthodontic group (P = .02).

There were no statistically significant differences in mean width, in mean greatest width, or in mean depth of sella turcica between the two groups.

Table 2 shows details on cephalometric analysis. Compared with the orthodontic group with bridging, the combined surgical-orthodontic group with bridging had a significantly greater mean SNA (P = .007), mean SNB (P = .012), mean lower face height (P = .0002), and mean maxillary mandibular plane angle (P = .006). The mean lower incisor to mandibular plane angle was reduced significantly in the combined surgical-orthodontic group (P = .009) compared with the orthodontics-only group.

This study compared the incidence of bridging and the dimensions of sella turcica in Caucasian subjects who had combined surgical-orthodontic correction of their malocclusion with a randomly selected Caucasian group who were treated contemporaneously by orthodontic means alone. The two parameters assessed in this study do not appear to have been compared previously in these two Caucasian groups. This study assessed bridging and sella turcica dimensions from lateral cephalometric radiographs, a means that has also been used in previous similar investigations on sella turcica.^4,9

The 7.3% incidence of bridging in the orthodontic group and the 16.7% incidence in the surgical-orthodontic group compare favorably with the 4.6–6% incidence and the 18.6% incidence reported, respectively, in a previous investigation.⁴ The reason for more type B bridging found in the combined surgical-orthodontic group is unclear although the ratio of type A to type B identified in our study is comparable with that found in a similar group by Becktor et al.⁴ From a total of 33 subjects in that study, 10 subjects had sella bridging of type A and 23 subjects had type B.

This study also demonstrated a statistically significant increase in the mean surface area and mean perimeter of the sella turcica in the surgical-orthodontic group compared with the orthodontics-only group. The mean interclinoid distance was significantly smaller in the surgical-orthodontic group, which concurs with the greater incidence of sella bridging in this group. Although enlargement of the sella turcica may be a sign of an intrasellar tumor or juxtasellar tumor,^11,12 asymptomatic enlargement of sella turcica may occur.¹³ Plain film radiographs have a relatively high sensitivity for detecting sella change at between 67% and 77% of positive findings,¹⁴ and clinicians should be suspicious when any of the sella turcica dimensions exceed the upper limits of normal.¹⁵ In subjects where an enlarged sella turcica was identified in this study, only one required further investigation.

In the subjects with bridging, no clinical symptoms were reported. This is in line with the findings of a previous study⁴ and may be due to a number of factors. Although fusion of the clinoid processes was identified radiographically, it may represent superimposition rather than bony union.⁴ Alternatively, these individuals may have been predisposed to sella bridging from a prenatal malformation involving a cartilage primordium¹⁶ or there may be an association with the path of the internal carotid artery.¹⁷ It has also been proposed that a sella turcica bridge and enlargement of the sella turcica may be the result of focal infections of the pituitary gland, which have not yet become clinically manifest.³

Statistically significant differences were found when comparing mean cephalometric values for the combined surgical-orthodontic group with bridging with those of the orthodontics-only group with bridging. The differences, with the exception of lower incisor to mandibular plane angle, all related to either anteroposterior or vertical skeletal parameters. This is not surprising because combined surgical-orthodontic treatment was used specifically to address skeletal disharmony.

The findings of this study indicate that sella turcica bridging was more than twice as common in Caucasian subjects who had combined surgical-orthodontic treatment compared with Caucasian subjects treated by orthodontic means alone. Significant differences in sella turcica dimensions also existed between these groups.

Sella turcica bridging was found to be twice as common in Caucasian subjects who had combined surgical-orthodontic treatment compared with Caucasian subjects who had orthodontics only. In Caucasians, the mean surface area and mean perimeter of the sella turcica were found to be significantly increased in subjects who had combined surgical-orthodontic treatment compared with those treated by orthodontic means only. The mean interclinoid distance was significantly smaller in the surgical-orthodontic group.

Table 1.Mean (SD) Sella Turcica Dimensions for the Combined Surgical-orthodontic Group (n = 150) and the Orthodontics-only Group (n = 150)

Table 2.Cephalometric Analysis (Mean and SD) for the Combined Surgical-orthodontic Group With Bridging (n = 25) Compared With the Orthodontics-only Group With Bridging (n = 11)

FIGURE 1. Dimensions of the sella turcica recorded

^aSpecialist Registrar, Unit of Orthodontics, Glasgow Dental School, Glasgow, UK