Axial Triangle of the Maxillary Sinus, and its Surgical Implication With the Position of Maxillary Sinus Septa During Sinus Floor Elevation: A CBCT Analysis

Sinus floor elevation (SFE) has become a common procedure for the rehabilitation of the edentulous posterior maxilla.^1,2  Surgical outcomes after sinus augmentation have become more predictable due to the introduction of cone beam computerized tomography (CBCT) in implant dentistry, along with accumulated knowledge regarding the influence of anatomical factors on the degree of surgical difficulty.^1,3–5  However, laterally approached sinus elevation can still be a burden to many clinicians. The thickness of the sinus membrane, intra-antral position of the posterior superior alveolar artery, and location and orientation of the sinus septa are major factors that dentists should consider before performing the procedure.^6–8  In this study, we suggest another anatomical consideration that could substantially affect access to and visibility of the surgical field.

The zygomaticomaxillary buttress (ZMB) is a bony ridge or eminence that is connected to the zygomatic process of the maxilla. The part of the ZMB that is involved in the SFE is a strong and prominent ridge extending upward from the socket of the first upper molar that creates a convex wall of the maxillary (Mx) sinus. Considering the anatomical position of the ZMB, the lateral wall of the sinus can be divided into anterior and posterior parts. Usually, the posterior part of the lateral wall falls into the so called “dark side.” It is often blocked from surgical view depending on the prominence of the ZMB and the toughness of the soft tissue of the patient's cheek (Figure 1).

For clinicians planning to perform SFE, especially if they are less experienced, it is helpful to know the exact anatomy of the sinus beforehand and anticipate surgical difficulties that might be encountered. The most problematic situation would be when surgeons need to avoid or control an existing antral bony septum, which could be addressed in several ways. However, limited access always makes the procedure much more difficult. Therefore, the location of the antral septa relevant to the ZMB may be one of the most important factors determining surgical difficulty. If the septa are located behind the ZMB, the SFE may be hindered by reduced visibility and accessibility.

Anatomical features of the Mx sinus have been investigated in many studies. According to a rare meta-analysis, the prevalence of sinus septa was 57.6%, and septa were found mostly in the first and second upper molar regions.⁹  Rancitelli et al¹⁰  reported that the prevalence of the sinus septa was 38.1%, and the medium thickness of the mucosa was thicker at the septal area. Schriber et al¹¹  found septa in half of the evaluated sinuses. Simsek Kaya et al¹²  evaluated posterior superior alveolar artery, sinus septa, accessory maxillary ostium, and lateral wall thickness. However, to the best of our knowledge, no study has focused on the prominence of ZMB or the location of sinus septa in relation to ZMB. Therefore, we suggest that the convexity of the lateral wall of the sinus is another anatomical consideration that should be thoroughly investigated preoperatively.

In this study, 2 angles were newly assigned from the axial view of CBCT images to determine convexity of the sinus lateral wall: the lateral sinus angle (LSA), the (anterior sinus angle (ASA), and the distribution of LSA and ASA in the Korean population. The locational relationship between ZMB and antral septa was also evaluated.

All data and images in this study were obtained from all patients who underwent SFE procedures for prosthetic restoration at the Department of Oral and Maxillofacial Surgery of Kyung Hee University Dental Hospital between January 2012 and June 2017. This study was designed as a cross-sectional study, and consecutive sampling was conducted. Every participant fulfilling the inclusion criteria and none of the exclusion criteria was analyzed using CBCT images, and the participants consisted of only those of Korean ethnicity in South Korea. This study was approved by the Ethics Committee of Kyung Hee University Dental Hospital (KHD IRB 1812-2).

Images were excluded if patients met the following criteria: (1) had sinus pathology, (2) had maxillary neoplasm, (3) were taking medication that would affect bone metabolism, (4) had previous sinus surgery, or (5) had a history of mandibular and maxillary fracture. This study was designed largely to analyze the outer shape of the Mx sinus somewhat distant from the alveolar bone compartment, where the shape was affected by pathologic conditions of the related tooth. Therefore, dental conditions such as dental caries, periodontal disease, periapical pathology, tooth loss, or any gingival inflammatory lesions were not considered in the exclusion criteria. To minimize the influence of aging, relatively stable anatomical landmarks (posterior nasal spine, zygomatic apophysis area of the maxillary bone) were used. However, dimensional changes of the Mx sinus might have influence.¹³  A total of 134 patients and 161 sinuses containing edentulous alveolar ridges were included in this study. The CBCT scans of all participants were taken using an Alphard 3030 Dental CT system (Asahi Roentgen Ind. Co, Ltd, Kyoto, Japan) according to a standard protocol (80 kVp, 5 mA, 17 seconds). The CBCT data were exported into DICOM format and imported into medical imaging software (OnDemand3D, Cybermed, Seoul, Korea) to analyze the images. Axial images were obtained by cuts made parallel to the Frankfort horizontal (FH) plane. A calibrated, experienced evaluator performed all the measurements, repeating them after at least 1 week to minimize measurement biases.

The sinus wall anterior to the ZMB was assigned as the anterior sinus wall, while the sinus wall posterior to ZMB was assigned as the lateral sinus wall. The LSA refers to an angle formed at the point of the identified ZMB between the anterior wall (anterior to ZMB) and the lateral wall of the Mx sinus (posterior to ZMB). The FH plane was chosen as the reference surface for image analysis of axial views. The FH plane refers to a cephalometric plane that passes through inferior borders of bony orbits and upper margins of the auditory meatus. It is commonly used as the reference surface in orthodontic diagnosis and treatment planning. When an axial view was selected at the level of the posterior nasal spine parallel to the FH plane, this cutting plane was positioned above the level of the sinus floor for all patients (Figure 1).

Since Mx sinus walls are not perfectly flat, 2 lines tangent to the anterior and lateral walls of the Mx sinus were created, and the angle between the 2 lines was measured (Figure 2a). The ASA refers to the angle formed between the midpalatal line and the anterior sinus wall. This location was also measured (Figure 2b). Consequently, the resulting calculation reviewed the convexity of the ZMB determined by the LSA and the ASA noted in the axial view of the CBCT image.

The prevalence of Mx sinus bony septa and their position relative to the ZMB were evaluated during this study. To determine factors influencing sinus elevation procedures, only the septa below the level of the nasal floor were included in this study. Therefore, coronally and sagittally oriented septa with a minimal expansion of 2 mm from sinus floor were recorded. The anteroposterior position of the antral septa and the ZMB (Figure 2) were recorded according to the location of their corresponding teeth. Teeth positions were defined by diagnostic stents manufactured for implant planning and representing future implant prosthesis. Subsequently, the relative position of the antral septa to the ZMB were evaluated.

Data collected from patient records according to the inclusion/exclusion criteria were statistically analyzed using IBM SPSS statistic software 22.0 (IBM Corp, Armonk, NY). A P value < .05 was used to assess the significance of statistical analyses. Continuous variables are presented as mean ± standard deviation, median, and range (Table 1) while categorical variables are presented as counts and percentages (Table 2). According to the Levene test, the homogeneity of variances for all independent groups was proven, and the parametric test assumptions were valid. Student t-test was performed to assess any difference in ASA between the left and right sides as well as between genders. The Student t-test was also used to compare LSA between the left and right sides as well as between genders. The methodology was reviewed by an independent statistician.

A total of 161 sinuses of 134 patients with a mean age of 56.6 years (median = 58 years, range = 19–90 years) were analyzed. There were 67 men and 67 women in the study, and 83 (51.5%) and 78 (48.5%) sinuses, respectively.

The LSA and ASA from CBCT measurements are shown in Table 1. Figures 3 and 4 present distribution of LSA and ASA, respectively. The mean LSA was 105.9° ± 9.86° (range = 76.9°–138.8°). There was no statistical difference between left and right sinuses (P = .420). The mean ASA was 58.4° ± 6.43° (range = 42.1°–77.6°). There was no significant difference between left and right sinuses (P = .564). There were no significant differences in LSA or ASA between men and women either (P = .912, P = 598, respectively).

Maxillary sinus septa located below the level of the nasal floor were found in 60 of 161 sinuses (37.3%), 53 sinuses had only 1 septum, while the rest of sinuses had 2 septa. The position of the septa was expressed as the location of their corresponding teeth (Figure 5). The second molar region was the most frequent (32.8%), followed by the first molar (20.9%), retromolar (16.4%), and second premolar regions (14.9%). The frequency in the interdental region (20.9%) was generally lower than that in the tooth region (70.1%) (Figure 5). The position of the septa relative to the ZMB was also examined. Table 2 shows the prevalence of septa according to their relationship with the ZMB. Septa were located mostly posteriorly to the ZMB (49.2%), followed by anteriorly (26.9%) and centrally (23.9%). Moreover, the ZMB was mostly located in the first molar region (66.4%), followed by the region behind the first molar (28.6%) and anterior to the first molar (5.0%).

One of the most important factors affecting the level of difficulty in surgery is the surgical approach, which is usually dependent on the visibility of the surgical field. For this reason, high convexity of the lateral wall provides challenges to surgeons surgically approaching the posterior region of the sinus, especially for reaching the second molar region. A convex lateral wall formed by a prominent zygomaticomaxillary buttress can conceal the posterior region. A steep anterior sinus wall also interferes with the surgeon's field of vision to the sinus wall. Since SFE is usually carried out under local anesthesia, excess retraction by the surgeon may increase frequent patient discomfort and soft tissue injury. If the antral septum is located behind the ZMB, the risk of surgical complications, such as perforation of the Schneiderian membrane during surgery, increases because of the difficulty in maneuvering membrane elevation in the presence of a septum under limited visibility. Therefore, this study also investigated the prevalence and position of sinus septa and their relationship with the ZMB.

Currently, CBCT is broadly used for preoperative evaluation of the Mx sinus. After its introduction to dentistry, many studies on anatomical structures of the sinus have been conducted. These investigations on posterior superior alveolar artery, sinus septa, and their dimensions^14–16  have helped oral surgeons overcome anatomical obstacles and obtain better surgical outcomes in SFE procedures. If anatomical obstacles are confirmed preoperatively, a modification of surgical technique and device, such as double window technique and piezoelectric device for antrostomy, can be suggested.^17,18  In the current study, CBCT images were instrumental. We believe these results can broaden insights into the Mx sinus.

The present study revealed that the LSA and ASA were normally distributed with average values of 105.9° and 58.4°, respectively. In 4.3 % of sinuses, LSAs below 90° were present, and this strong convexity was considered to be an obstacle for the procedure. Zijderveld et al⁶  reported that 6% of their patients undergoing SFE had a strong convexity of the lateral sinus wall, which led to development of an alternative method. Narrow ASAs indicate a steep anterior sinus wall, and it also restricts the surgeon's field of vision to the sinus wall.

The prevalence of sinus septa has been reported to be 20%– 57.6%. A single septum was more common than multiple septa.^9,10,19,20  The present study showed an incidence of 37.3% and revealed that 88.3% of sinuses had only 1 septum. In line with other studies,^3,9,11,15  septa were more frequently found in the molar region (62.7%), where implant placement would be more frequently required than in the premolar region (20.9%). More specifically, the prevalence was highest in the second molar region (32.8%). Interestingly, septa were located more in the tooth area (70.1 %) than the interdental and retromolar areas (20.9%). However, since the position of the tooth was estimated by a diagnostic stent, differences between the original position and estimated position might exist.

Although the convexity of the Mx sinus, including LSA and ASA, have not been described before, Niu et al²¹  have classified the Mx sinus contours into 5 categories on a coronal image: narrow tapered, tapering, ovoid, square, and irregular. It would be interesting to investigate the relationship between the Mx sinus contour on a coronal image and the convexity of the Mx sinus on an axial image.

This study also revealed that most septa were found in the region posterior to the ZMB (49.2%). These septa located in the area around the ZMB can complicate a sinus elevation procedure when an approach to the second molar is required. The ZMB was located in the first molar in 66.4% of sinuses as determined by a 3-dimensionally rendered image of the maxilla. The present radiographic study showed that the ZMB and the location of a septum might complicate SFE, and surgical expertise and experience are mandatory to successfully manage this type of clinical situation for optimum patient outcomes.

When a septum is observed during lateral wall preparation (Figure 6), the traditional surgical technique is to make 2 separate windows.²²  However, membrane elevation and creation of a cavity for bone graft may occasionally be complicated, because of limited surgical access to the posterior window. Under such conditions, a surgical modification is necessary to overcome this clinically demanding case. To ensure appropriate surgical access to the Schneiderian membrane and the sinus cavity, a septum may be resected as described in Figure 7. Surgical resection of a septum enabled 2 separate compartments to be united into a single large cavity, which is easier for the surgeon to fill during the procedure (Figures 8 and 9).

The convexity of the lateral wall of the Mx sinus and the ZMB has not been reported yet. Thus, the present study was performed to add more detailed knowledge and insight into the Mx sinus. A thorough preoperative assessment of the Mx sinus gives more precise information about SFE, and therefore, preparing surgeons to avoid and minimize complications. Since most septa are located behind the ZMB, as presented in this study, the convexity of the sinus wall and the ZMB should be carefully examined. In addition, a narrow ASA combined with a strong convexity could be more problematic. Because surgeons would have a restricted field of vision, there should be a more lateral approach to the surgical field. Hence, surgeons should prepare themselves with a proper preoperative evaluation and consider modifying the surgical technique, especially if septa are located behind the ZMB. However, the toughness of the patient's cheek soft tissue is also an important factor to determine surgical difficulty because visibility of the surgical field depends on it. Therefore, a surgeon's expertness in handling the soft tissue could be significant in maximizing visibility.

The present study used relatively stable anatomical landmarks (posterior nasal spine, the zygomatic apophysis area of the maxillary bone). However, individual anatomical variations of the landmarks might exist and could affect axial images selected at the level of the posterior nasal spine parallel to the FH plane and the measurement of LSAs and ASAs. In addition, the dimensional changes of the Mx sinus associated with aging might be influential. Results of this study should also be verified to determine if similar results could be found in other ethnic groups. It has been shown that northeast Asians exhibit wider malar regions than Caucasians,²³  meaning that Caucasians may have less prominent ZMBs. This indicates that the distribution of LSAs and ASAs may differ depending on ethnicity. Samples including various ethnicities are required to confirm these findings and generally apply them to preoperative evaluation of Mx sinuses. In addition, the calculation of sample size and statistical power should have been performed to ensure the reliability of the results. Therefore, the limitations of this study must be taken into consideration for future studies.

Despite these limitations, this study provides a novel approach to evaluate Mx sinus, that is, to measure the convexity of the lateral sinus wall and consider the positional relationship between ZMB and antral septa in cases when sinus septum is included in the surgical field. These considerations would provide additional information to prevent intraoperative complications, especially in demanding cases.

Abbreviations

ASA:

anterior sinus angle

CBCT:

cone beam computerized tomography

FH:

Frankfort horizontal

LSA:

lateral sinus angle

Mx:

maxillary

SFE:

sinus floor elevation

ZMB:

zygomaticomaxillary buttress

The authors are grateful to Professor J. I. Ryu for statistical support.

No conflicts of interest are declared.