Various papers have looked into MS morphometry and MS volumetric changes in relation to several conditions, i.e., different orthodontic appliances, nasal septum deviation, and sinus pathologies29,30,31. Koppe et al.32 took into consideration the MSV and the characteristics of the maxilla-facial skeleton of adults with an untreated bilateral cleft and negative control patients and reported that bigger skulls had greater MSs and that cleft patients had larger paranasal sinuses than the control subjects. On the contrary, Erdur et al.33 found that MSs of unilateral cleft patients had no statistically significant difference between sides. Other studies also evaluated the relationship between MS size with gender and age29,30,31.
Nevertheless, to the best of authors’ knowledge, the present study is the first human CBCT investigation to make a 3D analysis of the volume, surface and linear maximum width depth and height of the MS and to compare these data between three different skeletal classes, evaluating the relationship between the shape of the maxillary sinuses and the patients’ craniofacial features in adult subjects. Bidimensional orthopantomography and lateral cephalogram presents several shortcomings in assessing three dimensional structures as already investigated in literature. Every clinician with some experience with OPG knows that even in the same patient the surface of the maxillary sinuses may vary greatly because it is strictly dependent on the inclination of the head to the x-ray beam that causes different projection on the 2D sensor of two pyramid shape volumes with a particular orientation inside the maxilla.
All papers but one investigated the surface of maxillary sinuses on bidimensional radiographs therefore providing inaccurate data and making their conclusions almost worthless.
The only other study34 that investigated the three-dimensional volume of maxillary sinuses and its correlation to cephalometric indexes was published last year. Our study aimed at giving more information on correlations between cephalometric indexes and maxillary sinuses dimensions taking into account gender differences as it was not considered in the previous publication.
The data obtained in this study evidenced no significant difference for MSV, MSS, LMW, LMH and LMD between the right and left side, demonstrating the absence of asymmetry in the MS in the present sample.
A statistically significant difference was registered between genders for the MS with males having a larger sinus than females for each variable evaluated. One paper35 described that the maxillary sinus volume continued to grow until 20 years of age and then began reducing its size and their findings were in line with those in this study, in as much as they reported no significant difference between the right and left MSVs (Table 2).
The data of the present study showed no significant difference between the three skeletal class (ANB) group for the MSV and surface, evidencing no association between these variables and the anteroposterior relationship. There was a larger volume and surface in males than in females, in contrast with previously published literature data, i.e., Endo et al.22 reported having observed no statically significant differences in the maxillary morphological measurements between genders. Oktay’s study used the Duncan’s multiple comparison test to determine the size of maxillary sinuses and reported that that Class I female subjects and Class II and III male subjects had smaller MS than Class II female subjects1. The differences in these two results may be attributable to age dependent development of the MS.
Oktay analyzed 189 patients aged between 6 and 30 years, whereas Endo et al.22 included males and females of almost the same age. However, Oktay1 and Endo22 using bi-dimensional OPGs evaluated significantly distorted images of maxillary sinuses, with huge limitations in the methods which might well have hampered the interpretation of their findings, whilst the CBCT scans evaluated in this paper were significantly more accurate for MS three dimensional measurements.
Regarding linear measurements of the MS only LMW showed a statistically significant difference in ANB groups resulting larger in class I patients compared with class II and III (Table 3).
This is allegedly due to the association between class II and III malocclusion and maxillary hypoplasia that would reveal a reduced LMW. Various studies have reported that the anteroposterior and vertical problems related to Class II and III malocclusions are not the only related factors and that there is often an association with posterior transverse discrepancy36,37,38,39.
Class II and class III patients presented a reduced transversal dimension of the MS that had never been reported before in literature. This finding is strengthened by the Pearson’s correlation coefficients of the distance between the left and right Mx points with MSV, MSS and LMW that evidenced very high positive correlation (p-value < 0.01), as reported in Table 7.
Moreover, it was observed that females had a smaller LMW and LMH than males. No statistically significant interaction was noted for MS variables between gender and skeletal classes, with a similar trend for the MSV and MSS in the three SNA classes. This finding is inconsistent with those published by Oktay1 who evaluated the MS areas on OPGs of patients with different dental malocclusions defined according to Angle classification. Oktay1 found a significant interaction between gender and Angle’s class in the dimensions of the MS.
Differently from other previous research, the authors decided to make a 3D investigation assessing the influence that the cranio-maxillary relationship (SNA) may have on MS dimension. The authors supposed that the cranio-maxillary relationship (SNA) would have influenced maxillary sinuses size more than the sagittal skeletal jaw relationship (ANB).
The data of the present study showed that subjects with a retrusion (SNA < 80°) and protrusion (SNA > 84°) of the maxillary bone had a significant increase in MSV, MSS, LMW, LMH and LMD. Whilst there was no difference between genders for MSV, MSS and LMD. Conversely, males had larger LMW and LMH.
However, the data of the present study as to the relationship between SNA and MS dimensions cannot be compared to other research data as, to the best of authors’ knowledge, this is the first time this has been calculated. Interestingly, the evaluation of the significant interaction between the SNA classes and gender demonstrated a different tendency for the three linear variables between genders (Table 4).
The statistical interaction between SNA and LMW evidenced that females with a maxillary retrusion had smaller transversal sinus dimensions than males with the same condition. Conversely, males with maxillary protrusion (SNA > 84°) had smaller LMW than females.
The significant LMH findings on the interaction between SNA and gender evidenced that there was a different pattern for the two genders, i.e., in subjects with a reduced SNA, females presented smaller MS than males. A statistically significant interaction was found for LMD in subjects with normal SNA values. Whilst males presented significantly smaller LMD in subjects with normal SNA values compared with the other two subgroups, the female groups presented similar values for LMD in patients with normal or augmented SNA values and increased values of LMD in patients presenting reduced SNA values.
Several cephalometric measurements presented a positive correlation with the dimensions of the MS. These data allegedly sustain the hypothesis for the MS to be larger in patients presenting larger maxillofacial structures. Another study32 similarly reported that the MSV was positively correlated to the dimensions of the maxillofacial structures. This finding was in line with the data of the present study, even if Koppe et al.31 considered just three cephalometric parameters, that are the distance between the Mx points, facial height, and maxillary length.
The presented data did not evidenced any significant correlation between the SNA, SNB, and ANB angles and any MS size. This finding suggested that the anteroposterior dimension of the maxilla and of the mandible had neither a positive nor a negative correlation on MS size, although the ANB and SNA angles are both the criteria for the classification of the analyzed subjects (Table 7). These findings were in agreement with those previously published by Endo et al.21 who did not report any significant correlation between SNA, SNB and ANB angles and MS dimension. The abscence of a significant correlation between SNA and all the maxillary measurments is due to a V shape distribution of data between the SNA classes, with greater maxillary values in subjects with a retrusion and protusion of the maxillary bone.
Individuals with bigger S-N and PNS-A distances did tend to present larger MS. The correlation analysis of the MS measurements and the cranial base showed a positive correlation between S-N distance and MSV, MSS and LMD. Moreover, the significant positive correlation between the Ba-S-N angle and the LMD demonstrated that the angle increased along with the increase in depth. Conversely, the height and width were negatively correlated with the Ba-S-N angle. The data presented in this paper regarding Pearson’s correlation analysis between MS dimensions and PNS-A distance are in line with the one reported by Endo et al.22, that is subjects with longer nasomaxillary complexes have deeper and greater sinus, in terms of volume and surface. To understand this relationship we should consider that the MS, a pyramid-shaped sinus within the maxillary bone, is the biggest among paranasal sinuses40,41 as previously mentioned, contributing to midfacial growth and appearance22,42,43.
Endo et al.22 also reported a strong positive correlation between total maxillary sinus area (TMSA) and upper maxillary sinus area (UMSA) and S-N-Ans-Pns, showing a tendency for steeper maxillary planes to be positevely related with TMSA and UMSA. The results of the current research differ from those of Endo et al. as no relationship was observed between the inclination of the palatal plane (S-N/Ans-Pns) and the MS dimension and only the maximum height (LMH) had a statistically significant positive correlation with S-N-Ans-Pns.
The present correlation analysis between the vertical dimension and the MS measurements showed that patients with an augmented anterior vertical dimension should have larger MSV, MSS and greater LMH, and the subjects with a higher posterior facial height should have larger MSV, in accordance with data published by Endo et al.22. However, data comparison is not reliable as those previously published by Endo et al. were calculated on subjects aged from 12 to 16, a period when the maxillary size could be affected by a peak of growth.
Most of the studies are limited to the evaluation of the sinuses during orthodontic treatment or for a particular malocclusion and are impaired by bidimensional biases as well.
The present study aimed at obtaining information on the three-dimensional characteristics of maxillary sinuses amongst nasal breathing adults that had not been given any orthodontic treatment and evaluating the relationship that different cephalometric measurements may have on MSV and MSS, in both genders. However, this study does have some limitations, i.e., the relatively small sample size, even if it did suffice for inferential statistics consideration and the fact that all the subjects were Caucasian. Therefore, the conclusions may not be extrapolated to other ethnicities. Furthermore, the use of SNA and ANB are not completely reliable to investigate skeletal pattern. The literature teaches us that none of the sagittal parameters proposed so far can be considered totally reliable. It often happens that two parameters are conflicting or over or underestimate the actual skeletal discrepancy. The SNA and the ANB angle are parameters commonly used for evaluating the sagittal relationship between the maxilla and the cranial base and between the maxilla and the mandible respectively. The ANB angle would tend to decrease with increasing age44,45 and is influenced by the position of the nasion point: both on the sagittal plane and on the vertical plane46,47,48,49.
As reported by previous studies it seems that the growth of the MS in both genders overlaps with the pubertal growth spurt, and that its development starts between the age of 9 and 11 in females and later in males, between 12 and 1416. However, one strong point of this study is the average age of the selected sample, as it ensures a three-dimensional evaluation of the MS in an age range that is no longer subject to any significant volumetric changes. Moreover, the excellent intra-operator and inter-operator agreement (Table 7) demonstrated a very high reliability and repeatability of the presented method and a very small random error for volumetric and linear measurements in the assessment of MS size. The data of the present study are in line with various other authors who demonstrated the reliability and utility of CBCT scans in a three-dimensional assessment and reconstruction of the paranasal and maxillary sinus16,17,50.
The maxillary sinuses are of particular interest in dentistry due to their proximity to the area dentists work in. Therefore, thorough knowledge of MS anatomy is a must to avoid not only maxillofacial surgery complications, but also to make a presurgical evaluation for dental implant planning, graft size estimation for sinus lift procedures and infra-zigomatic mini-screw placement. Besides common dental procedure, the study of the MS is important in forensics to determine the gender if the whole body is not be available51. In case of orthodontic treatment that includes movements of the posterior teeth at the level of the maxillary sinus, the clinician should have a special care in patients with a large dimension of the maxillary sinus(as in case of an augmented vertical dimension). Literature reports that orthodontic space closure in missing posterior maxillary teeth through the sinus is arduous and it is recommended to use light force systems for a successful effect52,53. Moreover intrusion of teeth located in posterior area associated with root apices protruding into the maxillary sinus may be difficult and slow and it also need for very slight forces54,55.
The observations made in this study as to the orthodontic malocclusions seem to be in contrast with those put forward by Sassouni and Forrest56, i.e., that sinuses do not have a bearing on facial balance and malocclusions. A more thorough knowledge of three-dimensional MS dimension in different malocclusions may well be of help in treatment planning, as it has a common relationship with patients’ craniofacial features. However, all this considered and despite the efforts, the authors encourage these preliminary results be taken with caution as no clear-cut evidence is yet available in literature as to the influence cephalometric indices and/or malocclusive traits have on MS morphology. Moreover, even if present, statistical significance does not mean clinical relevance or the presence of cause/effect relationships between the variables. Indeed, statistical significance may have several confounding factors related to sample selection or sample pooling. Therefore, the authors urge the scientific community to carry out further studies on larger samples and to investigate further into any differences between populations to enrich and hopefully confirm the findings herein.
