INTRODUCTION

Aeration and drainage of frontal, anterior ethmoid and maxillary sinuses secretions have to follow a sinuous path through openings, fissures and grooves located on the lateral wall of the middle meatus. During development and ethmoid bone pneumatization, this region is anatomically prone to developing morphological variations of bone structures located there, such as agger nasi, uncinate process, ethmoid bulla, and middle nasal concha, modifying an already difficult space.

Even though many authors have related anatomical variations of the middle meatus considering predisposition to development of rhinosinusitis, there is no statistical or epidemiological evidence in the literature that such variations are the primary cause of rhinosinusitis, leading to significant controversy.

The contribution of anatomical variations in rhinosinusitis pathophysiology as an adjuvant in the maintenance cycle of the inflammatory process seems to be the most accepted theory.

A number of studies focusing on anatomical variations of the middle meatus region (ostiomeatal complex) analyzing computed tomography sections and of patients with clinical suspicion of recurrent acute and/or chronic rhinosinusitis can be found in the literature (Kennedy; Zinreich, 1988, Zinreich et al., 1988, Van Der Veken et al., 1990, Calhoun et al., 1991, Yousem; Kennedy, 1991, April et al., 1993, Zinreich et al., 1993, Nadas et al., 1995, Ünlü et al., 1994, Min et al., 1995, Nassar Filho et al., 2001 and Voegels et al., 20011-12), but nasal endoscopic studies are scarce.

Detailed nasal endoscopy enables the assessment of anatomical variations of the middle meatus, including agger nasi, uncinate process, ethmoid bulla and middle nasal concha. The proper knowledge of anatomical variations in this region provides more precise diagnosis and safer surgical access, as well as better understanding of the pathophysiology of nasosinusal infectious diseases.

The objectives of the present study were to assess the anatomical variations of the middle meatus of patients with nasal complaints using flexible videonasofibroscopy concerning (1) prevalence and main associations among them; (2) prevalence concerning gender and age of patients.

MATERIAL AND METHOD

Selection Criteria

In the present study, we analyzed the taped images recorded in videocassette tapes of 500 exams (flexible videonasofibroscopy) conducted at the Ambulatory of Otorhinolaryngology, University Hospital, Medical School, Federal University of Sergipe - UFS, during August 2000 to August 2001.

Of the 500 studied patients, 297 were female (59.4%) and 203 were male subjects (40.6%).

We excluded patients with history of nasal or paranasal surgery, facial trauma, presence of maxillofacial abnormalities, tumor or infectious process in activity.

Exam preparation and technique
Whenever possible and if there was no contraindication, we caused a lining mucosa retraction of the middle nasal conchae and of the nasal lateral wall by placing a cotton ball soaked in lidocaine 2% with vasoconstriction (adrenaline) diluted at 1: 120,000 in both nasal fossae and middle meatus for 5 minutes, which led to a non-painful and more detailed exam of the lateral wall of the middle meatus.

Assessment of anatomical variations
We established morphological definitions for different anatomical variations, naming them based on endoscopic findings and using anatomical parameters, rather than the analysis on CT scans data.

Based on the endoscopic exam, we named the anatomical variations of agger nasi as hypertrophied, morphologically characterized by significant medial projection of agger nasi cell with grooves delimiting and anterior-superior implant of the projected middle nasal concha, sometimes partially hidden or lateralized relative to the agger nasi (Figure 1).

According to the endoscopic images, we named and morphologically defined the anatomical variations of the uncinate process in:

 Non projected uncinate process, characterized by total absence of medial projection of the uncinate process on the nasal lateral wall, and we could visualize the natural ostium of the maxillary sinus and direct view of the ethmoid bulla, partial portion of the third lamella and retrobulbar recess (Figure 2).
 Angled or medially projected uncinate process, characterized by marked medial projection towards the middle nasal concha, narrowing the middle meatus (Figure 3).
 Perforated uncinate process, characterized by opening the ethmoid infundibulum and the nasal cavity (Figure 4).
 Curved uncinate process, characterized by an anterior curve of the anterior free margin, which may look as if it was a double middle nasal concha (Figure 5).

Based on the observation of the endoscopic exam, we named the anatomical variation of ethmoid bulla into hypertrophic, morphologically characterized by wide expansion of the ethmoid bulla in the middle meatus and towards the ethmoid infundibulum, narrowing the anatomical regions in addition to defining contact with the lateral wall of the middle nasal meatus, which could grow anteriorly and project outside of the middle meatus (Figure 6).

The variations of middle concha have also been observed in the endoscopic exam., and the following structures were named and morphologically defined as seen below:

 Globous middle nasal concha type lamellar (intralamellar cell) characterized by enlargement of the anterior-superior insertion of the middle nasal concha (Figure 7);
 Globous middle concha type bullosa, characterized by expansion of the middle nasal concha, restricted to the body, with bulbous morphology (Figure 8);
 Globous middle concha extensive type, characterized by marked expansion of the middle nasal concha, including its implant and significant narrowing of anterior passages to the middle meatus and the middle nasal concha and the nasal septum (Figure 9);
 Bi-lobe medium nasal concha, characterized by irregularity (depression) in the body of the middle concha inferior-posteriorly, giving the impression that there were two medial nasal conchae, one more anterior and the other more posterior (Figure 10);
 Grooved middle concha characterized by grooved depression longitudinally (anterior-posterior) in the body of the middle nasal concha (Figure 11).
 Paradoxical middle concha characterized by concavity directed to the septum and convexity directed to the ostiomeatal complex (Figure 12).

Endoscopic findings of anatomical variations mentioned above were compiled according to the examined side (right or left middle meatus, bilateral), age and gender.

Subjects were grouped into 3 age groups according to development and pneumatization of ethmoid. Group 1, up to 14 years of age (complete pneumatization); Group 2, aged 15 to 24 years (pneumatization still possible); Group 3, aged older than 25 years (terminated pneumatization).

The database was stored in microcomputer and processed using the Excel, so that we could study the anatomical variations and the main associations with gender, affected side and the different age groups.

Statistical analysis
The comparison of different anatomical variations and their main associations concerning the examined side, gender and different age groups was conducted using the chi-square statistical test. In cases in which chi-square was not applied, we used the exact Fisher's test. The level of confidence was 95% (p<0.05), as advocated for biological trials.

We conducted the proportion significance test to compare uni or bilateral anatomical variations as well as the comparison of the various age groups among each other.





RESULTS

We examined 500 patients or 1,000 middle meatus. Among them, 170 (34%) did not present any anatomical variation and 330 patients (66%) had at least one anatomical variation, regardless of the side.

Anatomical variations
We diagnosed 611 anatomical variations regardless of the side. Table 1 shows the distribution of unilateral and bilateral anatomical variations in the 500 studied patients. We also conducted the proportion significance test, comparing unilateral and bilateral (Table 1).

Main associations
Tables 2 and 3 show the figures of cases of the possible ipsilateral and contralateral associations of anatomical variations (two by two) and results of the statistical significance of the comparisons between the possible associations of anatomical variations using chi-square or Fisher's exact test.

Anatomical variations versus gender
Of the 203 male patients who were studied, 64.5% presented some level of anatomical variation (131 patients). Of the 297 female patients, 67% presented some type of anatomical variations (199 patients). We applied the chi-square test or Fisher's exact test for unilateral and bilateral anatomical variations, trying to observe whether there were differences concerning gender, but there was no statistically significant difference for the studied anatomical variations.

The incidence of anatomical variation in the different age group was 52.77% in Group 1, 50% in group G2 and 75.48% in group G3 (graph 1). As to significance among the groups, we obtained G1¹G2 (Zc= -2.26), G2¹G3 (Zc=-11.07) and G1¹G3 (Zc=-8.61), according to the proportion significance test showed (values outside the interval -1.96 to 1.96 rejected the hypothesis that the groups were the same, with confidence levels of 95%).






DISCUSSION

Endoscopy allows an excellent assessment of the mucosa that recovers the bone and cartilaginous structures that form the nasal fossa. Thus, the aspect of a structure in the endoscopic visualization should be correlated at some extent with the bone and/or cartilaginous structure by the mucosa.

Therefore, we established names and definitions based on the morphology of the anatomical variations.
In our study, we found 66% of patients with some anatomical variation, which agreed with the findings by Bolger et al. (1991) who found them in 64.9% upon analyzing computed tomography13.

Lloyd (1990) assessed agger nasi in CT scans and found agger nasi pneumatization or hypertrophy narrowing the frontal recess in 3% of the cases14. Voegels et al., 2001, also studied CT scans in patients with diagnosis of chronic rhinosinusitis and they found 1.07% of pneumatized agger nasi cells12. In the present study, we found hypertrophy of agger nasi in 17.2% of the patients. Even though the studies were conducted with different populations (orbital tumors and chronic rhinosinusitis), there was an overestimation of agger nasi pneumatization in the endoscopic exam compared to CT scan. It may be explained by the angle of vision of the endoscope, subjectivity of the examiner, or infectious process in activity, hindering the observation in the "CT scan window". Studies comparing CT scans and endoscopy findings in the same patients would help us understand such divergence of results.

Min et al. (1995) studied CT scans of patients with bilateral rhinosinusitis and they observed uncinate processes reserved in only 1.3% of the cases10. Through flexible nasofibroscopy we observed that the occurrence was only 0.02%. The low incidence are opposed to the findings by Lloyd (1990) who evidenced 16% of curved uncinate processes, but he considered in the classification all kinds of bending, even if they were slight, posteriorly from the margin of the uncinate process, medially towards the middle nasal concha, or laterally towards the ethmoid infundibulum 14. If we consider the curved uncinate process characterized by the free margin curved anteriorly giving the impression of a double middle nasal concha, this is not a very frequent abnormality and it is best diagnosed through nasal endoscopy than through CT scan, explaining the disparity of findings with Lloyd (1990).

Yousem, Kennedy (1991)5 and Zinreich (1993)7 mentioned the variation of the angled uncinate process or medially projected found in 3% to 21% of the CT scans of patients with suspicion of rhinosinusitis. Through endoscopy, we observed 9.2% of the cases, which is in accordance with the figures published in the literature.

In adults, Kennedy; Zinreich (1988) found 8% of the patients with hypertrophied ethmoidal bulla in the CT scan1, which corresponded to our findings (7.6%), reinforcing the efficiency of the endoscopic method in such diagnosis. Conversely, upon studying the CT scans of patients with orbital tumors, Lloyd (1990) found 17% of hypertrophy, but the author admitted that it could have been an overestimation since it is not always easy to assess such anatomical variation by the CT scan14.

The division of globous middle nasal concha into lamellar, bulbous and extensive was based on the classification used by Bolger et al. (1991), that differentiated pneumatized regions of the middle nasal concha through CT scan13. Upon analyzing coronal sections of CT scan, Bolger et al. (1991) found lamellar bulbous middle concha in 46.2% of the cases, bulbous type in 31.2% and extensive type in 15.7% of the patients13. In a similar study, Ünlü et al. (1994) observed 45% of lamellar type, 21% of bulbous and 34% of extensive type9. Conversely, in our endoscopic study we found a very reduced number of lamellar type (12.44%) and a high number of bulbous (43.98%) and extensive (51.03%) types. Such findings showed that there is a limitation in endoscopy to assess the pneumatization of the vertical lamina of the concha, which would correspond to the enlargement of the anterior-superior insertion of the middle nasal concha. Conversely, the overestimate of bulbous and extensive globous concha reflects again the easiness to observe them endoscopically on the anterior and inferior regions of the middle nasal concha and the fact that we included in globous conchae not only pneumatized concha but also hypertrophied one.

The tendency of having bilateral anatomical variations of the ethmoid was revealed in the endoscopic assessment of other structures such as the uncinate process projected or angled medially (48.26%), hypertrophied agger nasi (37.2%) and hypertrophied ethmoid bulla (31.58%), in addition to globous middle nasal concha of lamellar type (43.34%) and extensive (22.76%). Such behavior can represent a predisposition to uniform growth and pneumatization of the ethmoid bone sides.

We also observed a statistically significant difference between anatomical variations of hypertrophied agger nasi, hypertrophied bulla ethmoid, lamellar and extensive types of middle nasal conchae, paradoxical nasal concha and angled or medially projected uncinate process as bilateral occurrences (Table 3). The incidence of association between extensive middle nasal concha and bulbous concha contralaterally also presented statistically significant difference when compared to the other possible associations (Table 3).

We can infer that the growth and pneumatization of ethmoid bone tends to be uniform, since the anatomical variations referred above, except for paradoxical middle concha or angled or medially projected uncinate process, tend to be pneumatized.

Kennedy; Zinreich (1988), in a study with 230 adults, found 15% with paradoxical middle concha1. Lloyd (1990) observed them in 17%14, Bolger et al. (1991) in 26.1%13, Calhoun et al. (1991) in 12%4, Yousem; Kennedy (1991) in 2%5 and Voegels et al. (2001) in 2.15%12. Endoscopically, we found 9.8%, which is not a result of the limitation of endoscopy since this anatomical variation is easy identified by both methods (CT scan and flexible nasofibroscopy).

Upon analyzing Table 2, we observed a tendency of joint pneumatization of agger nasi and middle concha, extensive and bulbous types. We observed 29.06% of hypertrophied agger nasi associated with middle concha extensive or bulbous type, and 23.58% of bulbous middle concha and 20.32% of middle nasal concha extensive type associated with hypertrophied agger nasi. This finding was statistically significant when compared to other ipsilateral association found in the study. This relation was not a statistically significant difference between hypertrophied ethmoid bulla and hypertrophied agger nasi, and between hypertrophied ethmoid bulla and any type of globous middle nasal concha. The correlation of possible tendency to joint pneumatization of agger nasi (hypertrophied) and globous middle concha lamellar type was not noticed, maybe because we were working with endoscopic exams (flexible nasofibroscopy) and have not diagnosed the enlargement of anterior-superior implantation in some pneumatization situations which could have been detected by the CT scan.

There were no statistically significant differences of anatomical variations concerning gender.
Based on the study by Anderhuber; Khun (1993)15 who assessed the development of ethmoid pneumatization and concluded that the anatomical structures of the middle meatus had already been formed in the newborn leading to pneumatization of the ethmoid which is completed between the age of 12 and 14 years, and only in few cases there can be changes up to the age of 22 to 24 years, we divided our sample into three age groups. In Graph 1, we observe the statistically significant difference between presence (increasing) and absence (decreasing) of anatomical variations in the three age groups, as we advance in the examined age range.

This finding could suggest that the anatomical variations and ethmoid pneumatization develop progressively as the patient became older, but it could also result from greater tolerance to adult patients in conducting the exam, allowing better assessment of the middle meatus despite the strict methodology adopted by us during the flexible nasofibroscopy, excluding from the group of patients those could not be submitted to a detailed exam.

A longitudinal study, assessing the characteristics of the middle meatus of patients as a result of aging, could suggest a profile of follow up of anatomical variations, as well as to ethmoid pneumatization behavior.

After comparing and discussing the literature findings with CT scans and ours, through flexible endoscopy, we concluded that the methods complement each other, and CT scan is more precise for the diagnosis of pneumatization, but flexible nasofibroscopy seems to be better to assess morphology, as well as relative occupation of the space of the analyzed anatomical structures, giving the possibility of inferring about likely clinical consequences.

The studies comparing endoscopic and CT scan findings of the same patient and reassessing sequentially the same patient during his or her age development could elucidate many inferences made in the present study.

CONCLUSIONS

As to anatomical variations of the middle meatus in patients with nasal complaints assessed by flexible nasofibroscopy, we concluded that:

 Globous extensive and bulbous type middle conchae were the most frequent anatomical variations, followed by hypertrophied agger nasi, paradoxical middle concha and angled or medially projected uncinate process.
 Flexible nasofibroscopy tends to overestimate the level of pneumatization of nasal concha even though it fails to identify small pneumatization of the superior region.
 Agger nasi, extensive globous middle concha, paradoxical medial concha, angled or medially projected uncinate process, and hypertrophied ethmoid bulla tend to be bilateral, whereas globous bulbous middle concha tends to be unilateral.
 There was significant correlation between prevalence of hypertrophied agger nasi and extensive and bulbous globous middle concha ipsilaterally.
 There was statistically significant correlation between prevalence of extensive and bulbous type globous middle concha contralaterally.
 There was no statistically significant correlation between prevalence of anatomical variation of agger nasi, uncinate process, ethmoid bulla and middle nasal concha, isolated or associated, and gender.
 The prevalence of anatomical variations was significantly greater in the older age groups.

REFERENCES

1. Kennedy DW, Zinreich SJ. Functional endoscopic approach to inflammatory sinus disease. Current perspectives and technique modifications. Am J Rhinology 1988;2:89-96.
2. Zinreich SJ, Mattox DE, Kennedy DW, Chisholm HL, Diffley DM, Rosenbaum AE. Concha bullosa: CT evaluation. J Comput Assist Tomogr 1988;12:778-784.
3. Van Der Veken PJV, Clement PAR, Buisseret Th, Desprechins B, Kaufman L, Derde MP. CT-scan study of the incidence of sinus involvement and nasal anatomic variations in 196 children. Rhinology 1990;28:177-184.
4. Calhoun KH, Waggenspack GA, Simpson CB, Hokanson JA, Bailey BJ. CT Evaluation of the paranasal symptomatiz and asymptomatiz populations. Otolaryngol Head Neck Surg 1991;104:480-483.
5. Yousem DM, Kennedy DW, Rosenberg S. Ostiomeatal complex risk factors for sinusitis: CT evaluation. The Journal of Otolaryngology 1991;20(6):419-424.
6. April MM, Zinreich SJ, Baroody FM, Naclerio RM. Coronal CT scan abnormalities in children with chronic sinusitis. Laryngoscope 1993;103:985-990.
7. Zinreich SJ. Imaging of inflammatory sinus disease. Otolaryngol Clin North Am 1993;26:535-547.
8. Nadas S, Duvoisin B, Landry M, Schnyder P. Concha bullosa: frequency and appearances on CT and correlations with sinus disease in 308 patients with chronic sinusitis. Neuroradiology 1995;37:24-237.
9. Ünlü HH, Akyar S, Caylan R, Nalça Y. Concha bullosa. The Journal of Otolaryngology 1994;23(1):23-27.
10. Min Y, Koh T, Rhee C, Han M. Clinical implications of the uncinate process in paranasal sinusitis: radiologic evaluation. Am J Rhinology 1995;9:131-135.
11. Nassar Filho J, Anselmo-Lima WT, Santos AC. Participação das variações anatômicas do complexo ostiomeatal na gênese da rinossinusite crônica, analisadas por tomografia computadorizada. Revista Brasileira de Otorrinolaringologia 2001;67(4):489-495.
12. Voegels RL, Goto EY, Chung D, Nita LM, Lessa MM, Butugan O. Correlação etiológica entre variações anatômicas na tomografia computadorizada e a rinossinusite crônica. Revista Brasileira de Otorrinolaringologia 2001;67(4):507-510.
13. Bolger WE, Butzin CA., Parsons DS. Paranasal sinus bony anatomic variations and mucosal abnormalities: CT analysis for endoscopic sinus surgery. Laryngoscope 1991; 101:56-64.
14. Lloyd GAS. CT of the paranasal sinuses: study of control series in relation to endoscopic sinus surgery. The Journal of Laryngology and Otology 1990;104:477-481.
15. Anderhuber W, Kuhn F. Development of the paranasal sinuses in children: implications for paranasal sinus surgery. Ann Otol Rhinol Laryngol 1993;102:705-711.

---

1 Ph.D. in Medicine (ENT), University of São Paulo (USP).
2 Professor of Otorhinolaryngology, Federal University of Sergipe (UFS).
3 Graduated in Medicine, Federal University of Sergipe (UFS).
4 Resident Physician in Otorhinolaryngology, Federal University of Brasilia.
Address correspondence to: Avenida Gonçalo Prado Rollemberg, 211 Sala 412 Bairro São José Aracaju Sergipe 49010-410.
Tel (55 79)211-4576 - Fax (55 79)211-4517 - E-mail: isbo@infonet.com.br
Affiliation: Federal University of Sergipe - Department of Medicine, Federal University of Sergipe - Aracaju - Sergipe.
Ph.D. thesis submitted to the Medical School, University of São Paulo in March 2002.
Article submitted on December 09, 2002. Article accepted on March 13, 2003.