1

Introduction

Endoscopic frontal sinus surgery requires a detailed understanding of frontal sinus anatomy. The frontal recess and frontal sinus outflow tract vary greatly in their anatomy, often due to pneumatization of the frontal recess, which results in frontal recess cells. A normal variant in much of the population, frontal cells have the potential for causing frontal sinus disease, largely due to obstruction of the frontal sinus outflow tract . Because of the implications of these cells in frontal sinus disease and the various types of cells seen, a frontal sinus cell classification system was developed by Kuhn to allow prediction of prognosis and to facilitate communication regarding frontal recess anatomy . Frontal recess cells include the frontal cell types I-IV, agger nasi, supraorbital ethmoid, frontal bullar, suprabullar, and interfrontal sinus septal cell . A significant body of research has focused on the description and prevalence of frontal cells and their relationship to frontal disease. The presence of frontal cells has been associated with frontal sinus mucosal thickening, concha bullosa, and frontal sinusitis. Although research has established a connection between frontal cells and frontal sinus disease, the exact relationship between the prevalence of frontal sinus cells and incidence of frontal disease is unknown.

To our knowledge, no studies have been published specifically investigating gender differences and prevalence of frontal sinusitis. Ethnic differences in frontal sinusitis have been a focus of research, but no published research was found comparing African American and Caucasian populations. This study attempts to define the above relationships as well as the relationship between number and types of cells and disease to facilitate the identification and treatment of frontal sinus disease as it relates to gender, ethnicity, and frontal cell incidence.

2

Materials and methods

This study was performed at an academic medical center (University of Mississippi Medical Center in Jackson, Mississippi). This study was approved by the Institutional Review Board at the University of Mississippi Medical Center. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. We retrospectively identified all patients that underwent sinus computed tomography (CT) scans at the University of Mississippi Medical Center performed between the years 2003–2011. These scans were not limited to specific providers or specific diagnoses. Exclusion criteria included previous frontal sinus surgery, sinus malignancy, congenital anomalies, significant maxillofacial trauma, lack of axial, coronal, and sagittal CT scan views, and poor quality of scan. A total of 832 sinus CT scans were performed during the study time frame, and 602 scans met inclusion criteria. All CT scans had 0.6 mm axial cuts. Frontal sinus cell types I-IV, agger nasi cells, supraorbital ethmoid cells, suprabullar cells, frontal bullar cells, and interfrontal sinus septal cells were reviewed by the first author for each scan, right and left sides. The original frontal sinus cell classification system developed by Kuhn was used to define frontal sinus cells, with more detailed descriptions from recent publications ( Table A ).

The right and left sides of each scan were evaluated for evidence of frontal sinusitis. Frontal sinusitis was defined as > 3 mm of mucosal thickening involving the frontal sinus or the dependent portions of the frontal sinus, a commonly used definition based on previous studies.

Demographic comparisons of those with frontal sinus disease to those without frontal sinus disease were made with p -values computed according to Pearson’s chi-square statistic, with the exception of age, which was computed using a two sample t -test. A logistic regression, adjusted for ethnicity and gender, compared the proportion of patients exhibiting each type of cell between those with frontal sinus disease and those without frontal sinus disease. Frontal sinus cell comparisons were also made using a multivariable logistic regression model for ethnicity and gender as independent predictors of cell count. The total number of frontal sinus cells was compared between those with frontal sinus disease and those without frontal sinus disease.

2

Materials and methods

This study was performed at an academic medical center (University of Mississippi Medical Center in Jackson, Mississippi). This study was approved by the Institutional Review Board at the University of Mississippi Medical Center. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. We retrospectively identified all patients that underwent sinus computed tomography (CT) scans at the University of Mississippi Medical Center performed between the years 2003–2011. These scans were not limited to specific providers or specific diagnoses. Exclusion criteria included previous frontal sinus surgery, sinus malignancy, congenital anomalies, significant maxillofacial trauma, lack of axial, coronal, and sagittal CT scan views, and poor quality of scan. A total of 832 sinus CT scans were performed during the study time frame, and 602 scans met inclusion criteria. All CT scans had 0.6 mm axial cuts. Frontal sinus cell types I-IV, agger nasi cells, supraorbital ethmoid cells, suprabullar cells, frontal bullar cells, and interfrontal sinus septal cells were reviewed by the first author for each scan, right and left sides. The original frontal sinus cell classification system developed by Kuhn was used to define frontal sinus cells, with more detailed descriptions from recent publications ( Table A ).

The right and left sides of each scan were evaluated for evidence of frontal sinusitis. Frontal sinusitis was defined as > 3 mm of mucosal thickening involving the frontal sinus or the dependent portions of the frontal sinus, a commonly used definition based on previous studies.

Demographic comparisons of those with frontal sinus disease to those without frontal sinus disease were made with p -values computed according to Pearson’s chi-square statistic, with the exception of age, which was computed using a two sample t -test. A logistic regression, adjusted for ethnicity and gender, compared the proportion of patients exhibiting each type of cell between those with frontal sinus disease and those without frontal sinus disease. Frontal sinus cell comparisons were also made using a multivariable logistic regression model for ethnicity and gender as independent predictors of cell count. The total number of frontal sinus cells was compared between those with frontal sinus disease and those without frontal sinus disease.