Prevalence of Haller's Cells: A Panoramic Study.

CONTEXT: Haller's cells or infraorbital ethmoid cells refer to the ethmoidal pneumatization of the superior aspect of the maxillary sinus, and floor of the orbit can be seen on panoramic radiographs. They appear as well-defined, round-, oval-, or teardrop-shaped radiolucencies with smooth corticated or noncorticated borders. AIMS AND
OBJECTIVE: To determine the prevalence and scrutinize the characteristics of Haller's cells on panoramic radiographs. SETTINGS AND
DESIGN: The present study is a cross-sectional study carried out in institutional setting. SUBJECTS AND METHODS: The study group comprised 300 individuals with an age range of 08-80 years selected by convenient sampling method. Three hundred panoramic radiographs were examined for the presence of Haller's cells under ideal viewing conditions by two radiologists. STATISTICAL ANALYSIS USED: The data collected were tabulated and subjected to statistical analysis using SPSS version 20 for Windows (SPSS Inc., Chicago, IL, USA) to obtain the results. Chi-square test was performed, and significance was set at 0.05 levels.
RESULTS: Haller's cells were identified in 30/300 individuals giving a prevalence of 10% with 18 Haller's cells in males and 12 in females. Among the 30 Haller's cells, 14 were unilateral and 16 were bilateral. Most of the Haller's cells were oval followed by round shape.
CONCLUSION: The overall prevalence of Haller's cells is relatively low and is in harmony with other studies conducted in various geographic populations. Knowledge of Haller's cells while interpreting panoramic radiographs is essential to forewarn surgeons before endonasal procedures, thus preventing any untoward intraoperative complications.

INTRODUCTION

Haller's cells are defined as air cells situated beneath the ethmoid bulla along the roof of the maxillary sinus and the most inferior portion of the lamina papyracea, including air cells located within the ethmoid infundibulum.[1]

Albert von Haller, a Swiss anatomist, first described Haller's cells in 1765. These are also known as maxilloethmoidal or orbitoethmoidal cells.[2] Haller's cells are thought to arise in individuals with pneumatization of the lateral crus.[3] On imaging, these cells may be seen below the ethmoid bulla, along the line of maxillary sinus and the most inferior portion of the lamina orbitalis.[4] Haller cells are often seen on panoramic radiograph, and many authors have proven their prevalence in their studies. The importance of identifying Haller cells on panoramic radiograph is to rule out the patient's symptoms associated with this anatomical variation. These symptoms include orofacial pain and sinusitis, nasal obstruction, impaired nasal breathing, headache, chronic cough, and mucoceles.[5] Haller's cells can also cause difficulty to access the maxillary sinus or the anterior ethmoidal cells during endonasal procedures. Thus, identification of this anatomical variations is indispensable for the surgeon to avoid intraoperative complications.[2]

Infraorbital ethmoid cells have been described as well-defined, round-, oval-, or teardrop-shaped radiolucencies (single or multiple), unilocular or multilocular with a smooth border which may or may not appear corticated, and are located medial to the infraorbital foramen according to a solitary panoramic radiographic study.[6] The clinical significance of these entities and the lack of noteworthy research initiated this study.

Thus, this study was conducted with an aim to determine the prevalence of Haller's cells and with objective to scrutinize the characteristics of Haller's cells on panoramic radiographs.

SUBJECTS AND METHODS

Dental patients who are advised or referred for orthopantomograph, visiting the Department of Oral medicine and Radiology were selected by convenient sampling method in this study. Panoramic radiographs of the individuals were made using a panoramic unit (CS 8000C imaging software operated at 8–12 mA and 70–80 kVP) and were interpreted.

The study population comprised of 300 individuals in the age between 08 and 80 years. Patients with a history of trauma and/or surgery involving the maxillofacial region, systemic diseases affecting growth and development, or clinical and/or radiographic evidence of developmental anomalies/pathologies affecting the maxillofacial region were excluded from the study. The institutional ethical clearance was obtained before the conduct of the study. Following clinical examination, one panoramic radiograph for each of the patients was made.

The radiographs obtained were serially interpreted for the presence of Haller's cells under ideal viewing conditions. Two examiners examined the radiographs. Each examiner viewed the radiograph twice for the presence of cells in a time interval of 15 days. The presence of Haller's cell was confirmed only when interexaminer results matched. Interexaminer analysis revealed κ = 0.82, which represents almost perfect agreement.

The recognition of Haller's cells was made if an anatomical variation fulfilled the criteria suggested by Ahmad et al.[6] The observations pertaining to the Haller's cells were entered in the individuals’ pro formas. The data collected were tabulated and subjected to statistical analysis using SPSS version 20 for Windows (SPSS Inc., Chicago, IL, USA) to obtain the results. Significance was set at 0.05 levels. The Chi-square test was performed to analyze any significant difference between groups.

RESULTS

The mean age of the study participants was 31.28 years; 32.043 years in males and 29.182 years in females [Table 1].

Table 1

Number of study participants with the overall mean age and standard deviation

Gender	n	Mean	SD
Male	141	32.043	16.905
Female	159	29.182	14.880
Total	300	31.28	15.892

SD=Standard deviation

Out of 300 study participants, Haller's cells were found in 30, giving an overall prevalence of 10% [Graph 1]. The mean age for individuals with Haller's cells was 23.6 years.

Graph 1

Prevalence of Haller's cells

The individuals were divided into various groups according to the age as 8–18 years (Group 1), 19–28 years (Group 2), 29–38 years (Group 3), 39–48 years (Group 4), 49–58 years (Group 5), 59–68 years (Group 6), and 69–78 years (Group 7).

Out of 30 Haller's cells, 01 each were found in individuals of Groups IV, VI, and VII, 02 were found in individuals of Groups of III and V, and 05 were found in individuals of Group II while 18 were found in Group I. P < 0.01 rendering it significant which states that Haller's cells are less evident as the age advances [Table 2].

Table 2

Distribution of Haller’s cells with respect to age

Age groups	Haller’s cells	Total	χ2	P
8-18 (Group 1)	18	72	24.683	P < 0.01
19-28 (Group 2)	5	124
29-38 (Group 3)	2	34
39-48 (Group 4)	1	24
49-58 (Group 5)	2	19
59-68 (Group 6)	1	15
69-78 (Group 7)	1	12
Total	30	300

Out of 30 Haller's cells, 18 were found in males and 12 were found in females, thus making male to female ratio of the individuals with Haller's cells as 3:2.

The mean age for males with Haller's cells was 25.55 years, and for females, it was 20.66 years.

Among 30 Haller's cells, 14 were unilateral and 16 were bilateral.

Among 14 unilateral Haller's cells, 11 were seen in males and 03 in females, and 10 were localized on the left side and 04 were localized on the right side. Among the 16 bilateral Haller's cells, 07 were found in males and 09 were seen in females [Graph 2 and Table 3].

Graph 2

Distribution of Haller's cells according to site

Table 3

Distribution of Haller’s cells with respect to gender and laterality

	Laterality		Total	χ2	P
	Unilateral	Bilateral
Gender
Male	11	7	18	3.772	0.072
Female	3	9	12
Total	14	16	30

Among the 30 Haller's cells, 28 were of unilocular type and only two were multilocular type. Among the 14 unilateral Haller's cells, 10 were found on the left side in which eight were unilocular type and one multilocular type and one individual had both unilocular and multilocular and all four on the right side were unilocular type. Among 16 bilateral, 14 were unilocular type and one multilocular and one individual had both unilocular and multilocular type. Moreover, among 30 Haller's cells, 29 are corticated and one individual is noncorticated on the left side [Table 4].

Table 4

Distribution of Haller’s cell according to site and locularity and cortication

Site	Cortication				Total	χ2	P
	Unilocular	Multilocular	Both	Noncorticated
Right	4	0	0	0	4	3.255	0.776 (NS)
Left	7	1	1	1	10
Bilateral	14	1	1	0	16
Total	25	2	2	1	30

NS=Not significant

It was seen that 53.33% of cells were oval in shape followed by round cells (46.66%) [Table 5].

Table 5

Distribution of Haller’s cell according to shape

Shape	n (%)
Pyramidal	2 (6.66)
Heart	6 (20)
Round	14 (46.66)
Oval	16 (53.33)
Triangular	8 (26.66)
Teardrop	3 (10)
Multiple (more than one shape)	13 (43.33)

Maximum number of Haller's cells was in the age group of 8–18 years. However, no significant correlation was observed between the age groups and shape of the cells (P = 0.819).

Furthermore, no significant correlation was observed between the gender and shape of the cells (P = 0.311).

A significant correlation was observed when the shape of the cells and side of cells were correlated, P = 0.027. Of 16 individuals with bilateral presentation, 11 individuals had multiple (more than one shape) Haller's cells, which was found to be statistically significant [Table 6].

Table 6

Distribution of Haller’s cell according to shape and side

Shape	Site			Total	χ2	P
	Right	Left	Bilateral
Pyramidal	0	1	0	1	20.234	0.027 (S)
Round	1	3	0	4
Oval	2	3	3	8
Triangular	0	1	2	3
Teardrop	1	0	0	1
Multiple (more than one shape)	0	2	11	13
Total	4	10	16	30

S=Significant

Of 14 individuals with unilateral presentation of Haller's cells, four had a round and five had an oval presentation. Only three individuals showed pyramidal-, triangular-, and teardrop-shaped Haller's cells and two individuals showed multiple (more than one shape) Haller's cells [Figure 1]. A significant difference was observed between groups (P = 0.001).

Figure 1

Panoramic radiograph showing right side heart shape, round, triangular Haller's cells and left side teardrop, round, pyramidal shape Haller's cells

Furthermore, no significant correlation was observed when the shape of the cells and loculae were correlated (P = 0.567).

DISCUSSION

Ethmoidal sinus development starts in the 10th week of intrauterine life and is radiographically visible by six months of age. The ethmoidal sinus grows rapidly in the first few years of life and attains full size by 12–14 years of age. During development, these ethmoid air cells tend to expand and occupy all the available space resulting in the formation of extramural cells. Seydel called this phenomenon “the struggle for space of the ethmoid.” Hence, these Haller's cells originally belong to the family derived from the extramural migration of the ethmoidal sinus into the floor of the orbit. Concha media bullosa is another variant of extramural cells due to pneumatization of the anterior middle turbinate through the posterior ethmoidal air cells.[7]

The present study sample consisted of 300 individuals in the age between eight and 78 years. The maximum number of individuals in the present study were from the age group of 19–28 years (Group II) followed by those from below 18 years age group (Group I) and the number decreases further with increasing age group. The reason is that the patients from these particular age groups visit college mostly for orthodontic treatment and third molar extraction, for which panoramic radiograph is recommended. The individuals in the present study were mostly females may be due to more concern of females for esthetic purposes.

A wide range in the prevalence (4.7%–45.1%) of infraorbital ethmoid cells has been reported in the literature. In the present study, we found a prevalence of 10% (30 out of 300 individuals), which fell within the range of previous studies. A much higher prevalence of 38.2% was reported by Ahmad et al.[6]

Raina et al. reported a prevalence of 16%, Solanki et al. reported a prevalence of 19.2%, Ramaswamy et al. showed the prevalence of 20.75%, Khayam et al. observed a prevalence of 32%, and Davoodi showed the prevalence of 17.8% in their studies.[25789] This difference could have been to variation in the population studied and the sample sizes and the subjective judgment pertaining to the presence or absence of Haller's cells.

Out of 30 Haller's cells, 18 were found in males and 12 were found in females, thus making male to female ratio of the individuals with Haller's cells as 3:2. This is in agreement with the study conducted by Raina et al. and Khayam et al.[28] A higher prevalence in females was seen in the study by Ahmad et al., Ramaswamy et al., Davoodi et al., and Khan et al.[67910] This variation in the male to female ratio could be due to recruitment of individuals in the study, irrespective of the gender. Contrary to this, no significant difference was reported among prevalence of Haller's cells between males and females in studies conducted by Basić et al. and Solanki et al.[15]

The overall mean age for individuals with Haller's cells was 23.6 years. The mean age for male individuals with Haller's cells was 25.55 years, and for females, it was 20.66 years. Dissimilar results were shown by other studies where mean age of individuals with Haller's cells was more than the present study.[27810] This variation may be due to recruitment of individuals in their study.

The individuals were divided into various groups according to the age as 8–18 years (Group 1), 19–28 years (Group 2), 29–38 years (Group 3), 39–48 years (Group 4), 49–58 years (Group 5), 59–68 years (Group 6), and 69–78 years (Group 7).

Out of 30 Haller's cells, one each was found in individuals of Groups IV, VI, and VII, 02 were found in individuals of Groups of III and V, and 05 were found in individuals of Group II while 18 were found in Group I individuals.

This is in accordance with previous studies which showed maximum number of Haller's cells in age group of 18–28 years.[257891011] The number of Haller's cells are less from age group of 39–78 years which is in agreement with previous studies.[258910] The similarities and differences shown in number of Haller's cells may be due to higher number of individuals pertaining to that particular age group in their study.

The total number of unilateral Haller's cells in the present study was 14 and bilateral Haller's cells was 16, suggestive of that Haller's cells mostly occur bilaterally. This is in favor of study by Khan et al. where bilateral cells are dominant than unilateral.[10] This is in contrary with the previous studies wherein Haller's cells mostly occur unilaterally than bilaterally.[2571112] No significant difference is noted in unilateral and bilateral cells in study by Khayam et al.[813]

Among the 14 unilateral Haller's cells, 11 occurred in males and three occurred in females. Among the 16 bilateral Haller's cells, seven occurred in males and nine occurred in female. Only three studies have been done in the past comparing laterality and gender.[578]

Among the 14 unilateral Haller's cells, 10 were localized on the left side and four were localized on the right side. This is in accordance with study by Khan et al.[10] No statistically significant differences were noted in the occurrence of Haller's cells on the right and left side in study conducted by Raina et al. and Solanki et al. and Khayam et al.[258] This variation could have been due to racial variation in the sample size as well as the population studied.

Out of 30 Haller's cells, 28 were of unilocular type and only two were multilocular type, making the unilocular type the most common type. Among the 14 unilateral Haller's cells, 10 were found on the left side in which eight were unilocular type and one multilocular type and one individual had both unilocular and multilocular, and all four on the right side were unilocular type. Moreover, among 30 Haller's cells, 29 are corticated and one individual is noncorticated on the left side. This is the first study to give correlation of Haller's cell according to site and locularity and cortication.

Among the 30 Haller's cells, 28 were unilocular and only two were multilocular, making the unilocular type the most common type. Similar results were obtained by the study of Raina et al., Solanki et al., and Khan et al. where unilocular type is more than multilocular.[2510] The discrepancy of the type of Haller's cells might be because of population variation, variation in the sample size as well as due to interobserver variability in determining the type of Haller's cells.

In present study, of 62 Haller's cells, the majority of the Haller's cells were oval or round in shape with very few cases depicting a teardrop shape [Figures 2 and 3]. This is in accordance with previous studies.[25710] Two new shapes have been seen in our study of pyramidal- and heart-shaped cells, which has been in study by Solanki et al.[5] One new shape – triangular had been reported in our study which has not been seen in any study previously.

Figure 2

Panoramic radiograph showing right side round-shaped Haller's cells and left side heart-shaped Haller's cells

Figure 3

Panoramic radiograph showing left side oval-shaped Haller's cells

The results of several studies emphasize to the clinical importance of Haller's cells because even if infraorbital ethmoid cells are not diseased, their presence may narrow the ethmoid infundibulum or the ostium of the maxillary sinus.[6]

Such anatomic limitation is seen constantly with rhinosinusitis.[14] Anatomic obstruction of the infundibulum with the presence of huge Haller's cells can cause blockage in the transmission of fluids.

Alkire and Bhattacharyya evaluated the effects of septum deviation, concha bullosa, and Haller's cells on the occurrence of acute rhinosinusitis, and their results showed that just obstruction caused by Haller's cells can lead to the disease.[15] A review article has also reported about headache related to Haller's cells,[16] and it has been said that Haller cells may also cause sinus disease such as mucocele.[17]

Sebrechts et al. acknowledged that Haller cell inflammation can be a potential reason of orbital unilateral edema.[18]

On the other hand, some studies suggested that the presence of Haller's cells automatically does not predispose an individual to the sinus disease.[1920] Ahmad et al. also did not report any symptoms of these diseases.[6] In our study, exclusively radiographic evidence has been studied, and pathological problems and symptoms associated with these cells have not been evaluated.

The ability to identify these cells on routine panoramic radiographs can help come to diagnosis of many conditions. Use of advanced imaging modalities would further help in confirming the presence of cells and formulating a definitive diagnosis related to patients’ symptoms.

CONCLUSION

The overall prevalence of Haller's cells is relatively low which is in harmony with most of the similar studies conducted on various geographic populations. The results of this study indicate that panoramic radiographs can depict and provide a clear delineation of Haller's cells. Knowledge of Haller's cells may be helpful in interpreting panoramic radiographs which may also forewarn surgeons before endonasal procedures, thus preventing any untoward intraoperative complications.

Our study highlights one new shape of Haller's cells other than the various parameters recorded to distinguish these cells. This shape has not been reported in the previous studies. The findings in the present study are purely confined to the extent of morphological characteristics observed on two-dimensional imaging modalities such as panoramic radiographs. Further studies employing advanced imaging modalities would aid in justifying our findings and providing a more precise description of these less explored entities.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.