Computed Tomography Assessment of Nasal Septal Mucosa: Establishing Criteria for Thickness Classification

Introduction

The nasal septum is composed of cartilage anteriorly and bone posteriorly, and it is enveloped anteriorly on both sides by the nasal septal mucoperichondrial flap (NSMF). The anatomical structure and surgical manipulation of the NSMF are critical in surgeries involving the nasal septum, such as septoplasty. Variability in the thickness of the NSMF is a crucial factor that influences outcomes in these procedures. A thinner NSMF predisposes patients to mucosal tears during surgery, leading to potential complications such as septal perforations. This study aims to assess the correlation between preoperative computed tomography (CT) measurements of septum thickness and intraoperative findings, explore demographic influences on mucosal thickness, and examine how these factors correlate with surgical outcomes, particularly the incidence of mucosal tears. The insights gained will be invaluable for enhancing surgical techniques and improving patient outcomes.^1,2

Despite recognizing the clinical relevance of NSMF thickness, the literature lacks a sufficient number of studies addressing this concern. This gap underscores the necessity of a reliable method to assess and classify septum thickness preoperatively. This research seeks to establish a radiological classification for septum thickness using CT scans, correlating it with surgical observations of NSMF thickness to enhance clinical applicability. Prior studies, such as a cadaveric one, have measured septal thickness but lacked practical clinical interpretation. ³

The study hypothesizes that a thinner NSMF correlates with an increased risk of intraoperative mucosal tears. By classifying the septum into “thin” and “thick” categories based on preoperative CT measurements, we aim to provide a predictive tool for surgical complications, thereby facilitating better preoperative planning and patient counseling. The classification system derived from this study could revolutionize the surgical approach to the nasal septum, improving common procedures such as septoplasty. ⁴

Materials and Methods

Study Design

Septoplasty was performed under general anesthesia with an endoscopic approach. We administered 2 ml of 1% lidocaine with 1:100,000 epinephrine to each side of the nasal septum. A nasal pledget saturated with a 1% xylometazoline solution was placed in each nostril. The surgery commenced with an incision, utilizing the Killian technique; however, if the deviation was anterior, a hemitransfixion incision was the preferred choice. For enhanced visualization, we employed a 4 mm rigid endoscope from Karl Storz. Following the procedure, bilateral intranasal Doyle silicone splints were then inserted, they are typically removed between the fifth and seventh days. We then scheduled follow-up appointments 3 weeks after surgery, with additional visits as required if sinus surgery has been performed.

Furthermore, we used standard software to perform radiological CT scan imaging to assess the nasal septum thickness. CT scans were performed using either the GE Healthcare Revelation, capable of 256 slices with Adaptive Statistical Iterative Reconstruction (ASiR) technology for dose optimization, or the Siemens Somatom Definition AS, which offers up to 128 slices and includes Care Dose 4D technology for dynamic dose adjustment. Measurements were taken at the Kilian incision area and saved on an Excel spreadsheet. The imaging protocol included measuring septum thickness at specific anatomical landmarks. Measurements were taken on the coronal plane from the anterior tip of the nasal bone to the nasal spine (Figure 1A) and on the sagittal plane where the lower third intersects with the upper two-thirds (Figure 1B). Utilizing the “ruler” function on the coronal plane (Figure 1C), measurements were recorded in millimeters and rounded to the nearest tenth for accuracy. An independent radiologist, FMA, performed these measurements following a standardization session with surgeon AA to ensure consistency. This approach aimed to align the radiological measurements with the precise locations of incisions, thereby enhancing the relevance of the data to surgical outcomes.

OPEN IN VIEWER

Figure 1.

Illustration of the method for determining septal thickness in CT scans. CT measurement technique. This figure illustrates the method for determining septal thickness in CT scans, which is integral to preoperative assessment and planning. (A) The plane is anterior to the nasal bone and anterior nasal spine. (B) The juncture where the lower third intersects with the upper two-thirds on the sagittal plane. (C) Recorded measurement on the coronal plane where (A) and (B) intersect (red line).

Variables

The variables collected and included in the study are as follows:

Date of Birth

Age at the time of surgery

Gender

Date of Surgery

Other variables: collected encompassed the presence of primary sinus surgery concurrent with septoplasty, the configuration of nasal septal deviation, the presence of caudal dislocation, the site of mucosal incision, and the occurrence of septal tears intraoperatively.

The classification of NSMF thickness into “thin” or “thick” categories was determined based on intraoperative assessments and documented in the patients’ standard operative records. In Figure 2, an example of thin (Figure 2A) and thick (Figure 2B) as subjectively reported intraoperatively. Measurements of nasal septal thickness on CT scans were conducted blind to the intraoperative evaluations to minimize bias, ensuring that the person measuring the CT thickness was unaware of the intraoperative thickness assessments.

OPEN IN VIEWER

Figure 2.

An example of an intraoperative endoscopic view subjectively described as thin (A) versus thick (B) nasal septal mucosa.

Results

Using normality tests (Shapiro-Wilk test), we established that CT thickness and age did not follow a normal distribution (CT thickness P < .001; Age P = .006). Spearman rank correlation was used to assess the relationship between age and CT thickness. For our independent testing, we used the Mann-Whitney U test. The distribution of intraoperative thickness assessments is depicted in Figure 3.

OPEN IN VIEWER

Figure 3.

Distribution curve of measured septal thickness. The distribution curves depict the overlap of CT-measured thickness for the categorized “thin” and “thick” groups as per subjective intraoperative evaluation. CT, computed tomography.

ROC Curve Analysis

Figure 4 shows a highly significant area under the curve for CT thickness [AUC = 0.884, se = .025, P < .001, 95% CI (0.835, 0.933)]. This confirms that a 3.65 mm cutoff can effectively classify NSMF thickness with a sensitivity of 80.9% and specificity of 78.5%. The PPV is 54.29%, while the NPV is high at 92.86%, thus reliable detection of “thin” mucosa in surgery. Subjective assessments had an accuracy of 79.08% compared to CT scans.

OPEN IN VIEWER

Figure 4.

ROC analysis. The ROC curve demonstrates the diagnostic ability of CT-measured septal thickness to differentiate between “thin” and “thick” nasal septal mucoperichondrial flaps using the established cutoff value. CT, computed tomography. ROC, receiver operating characteristic.

Logistic Regression

Logistic regression identified age as a predictor for tear occurrence, with each year increasing the odds of no tear by 6.6% (P = .001, OR = 0.934), and the model’s overall fit is evaluated using the −2 log-likelihood, Cox & Snell, and Nagelkerke R Square values, alongside the Hosmer and Lemeshow Test, with a chi-square of 13.370 and a significance of .100, suggesting an adequate fit. The classification table shows the model’s predictive performance, with an overall percentage correct of 66.8%.

Table 1 delineates the comparative data on age and CT thickness pertaining to the “Thin” versus “Thick” classifications based on intraoperative assessments. Table 2 details the analysis of variables by CT Thickness Cut-off and statistical significance.

OPEN IN VIEWER

Table 1.

Comparative Data on Patient Age, Gender, and CT-Measured Septal Thickness.

	All participants	Intraoperative subjective reporting of nasal thickness
		Thin	Thick	P value
Age in years (average, SD)	31.99 (8.8)	32.5 (9)	30.4 (8.1)	0.126
n (%)	196 (100)	149 (76)	47 (24)
Measured CT thickness (mm) (Average, SD)	3.45 (0.89)	3.15(0.7)	4.40 (0.9)	<0.001
n (%)	196 (100)	149 (76)	47 (24)
Gender, male	137 (69.9)			0.957
n (%)		104 (69.8)	33 (70.2)

This table presents comparative data on patient age, gender, and CT-measured septal thickness, categorized according to subjective intraoperative assessments of “thin” and “thick” nasal mucosa.

Abbreviation: CT, computed tomography.

OPEN IN VIEWER

Table 2.

Analysis of Variables by CT Thickness Cut-off.

The cut-off value of 3.65 mm		Below	Above	P value
Age in years (average, SD)		32.16 (8.8)	31.68 (8.9)	0.577
Gender, n	Male	79	58	0.003
	Female	47	12
Was sinus surgery done in addition to septoplasty?	Yes	65	39	0.579
	No	61	31
How do you describe the deviated nasal septum?	Deviated to right	60	26	0.219
	Deviated to left	62	39
	Complex deviation	4	5
Is there a caudal dislocation?	Yes	4	4	0.459
	No	122	66
Tear in the flap?	Yes	89	47	0.611
	No	37	23

Abbreviation: CT, computed tomography.

Discussion

The study findings indicate that using CT scans to measure septum thickness correlates well with intraoperative observations. A threshold of 3.65 mm effectively distinguishes between thin and thick flaps, exhibiting 80.9% sensitivity and 78.5% specificity, achieving an overall accuracy of 79.08%. These results highlight the utility of high-resolution CT imaging in preoperative assessments. However, our comprehensive literature review did not reveal any prior studies that categorize nasal septum thickness based on CT findings.

In a cadaver study utilizing Techlock GM-21 Calipergauze, the septal thickness was measured post-making parasagittal incisions to expose the septum. Variability in thickness was observed across different regions: the base anterior to the vomer was the thickest area, whereas a section just above the base was the thinnest. Measurements showed that the anterior-inferior region above the thinnest part was thinner compared to the superior-posterior part, indicating a gradient in septal cartilage thickness throughout the nasal septum. ⁵

The septal cartilage thickness in previous studies was observed to vary between 0.8 mm and 2.2 mm. Similarly, the mucosal thickness ranged significantly, from 0.3 mm to 3.7 mm, indicating a notable variability in these measurements across different parts of the nasal septum. ⁶ In the study by Thamboo et al., ⁷ thickness measurements were specifically taken at the thickest soft tissue areas on one side of the septum mucosa in various nasal regions. Recorded measurements showed an average anterior septum thickness of 3.28 mm (SD 0.13 mm) in control subjects, 3.0 mm (SD 0.13 mm) for Empty Nose Syndrome (ENS) patients, and 3.04 mm (SD 0.16 mm) for those with past inferior turbinate reduction without ENS, with no significant statistical difference between these groups.

In discussing the results of this study, we reflect on the alignment of subjective intraoperative assessments with objective CT measurements of nasal septum thickness. The study finds no significant difference in the age at surgery when comparing “Thin” versus “Thick” mucosa, while a significant difference in CT thickness between these groups is noted. Gender shows no significant influence on the subjective reporting of nasal thickness, and tear occurrence shows no correlation with subjective nasal thickness assessments. Notably, males exhibit thicker septum than females on CT.

Age differences between males and females at surgery are insignificant, while FESS with septoplasty is more common in older patients. A related study on complications in endoscopic sinus surgery patients identified the average age to be approximately 36 years. ⁸ Tear presence shows no significant correlation with CT thickness. The study also illustrates procedural trends with the majority of incisions on the left and the prevalence of small unilateral flap tears. Furthermore, the study introduces a standardized 3.65 mm cutoff for CT thickness classification with high sensitivity, specificity, and accuracy.

The significant association between gender and CT thickness classification after the 3.65 mm cutoff is discussed, revealing gender disparities in nasal septum thickness. Literature is lacking studies that directly compare nasal mucosa thickness between genders, making our findings particularly significant. Based on the results provided, females are more likely to have a “thin” nasal septum when using a cutoff value of 3.65 mm for septal thickness. The data show that out of 59 females, 47 (approximately 79.7%) fall under the “Below” category for thin septal mucosa, compared to 79 out of 137 males (approximately 57.7%) being classified as “Below.” Logistic regression highlights age as a significant predictor for the occurrence of tears during surgery, emphasizing the importance of age in preoperative planning. Logistic regression showed that each additional year decreases the risk of mucosal tears by 6.6%. This finding might seem counterintuitive since mucosal atrophy typically increases with age. However, the literature review reveals that age was not a significant risk factor for septal perforations in a study ¹ involving 143 patients undergoing septoplasty or septorhinoplasty, where age did not differ significantly between those with or without perforations (P = .665). While not all mucosal tears lead to permanent septal perforations, further research is needed to thoroughly examine the impact of age on tear and perforation risk.

The ROC curve analysis revealed an AUC of 0.884, demonstrating the efficacy of CT in discriminating mucosal thickness. The established threshold of 3.65 mm is corroborated by its notable sensitivity and specificity, contributing to its clinical applicability. A high NPV underscores its effectiveness in correctly identifying “thin” mucosa, an essential clinical advantage. Conversely, the PPV suggests potential enhancements in detecting “thick” mucosa. Overall, the reliability of CT for preoperative evaluation is affirmed, although optimization could further its precision.

The discussion is enhanced by data visualizations, including ROC curves, which substantiate these findings. In addition, it acknowledges the necessity for continued research aimed at improving the predictability and prevention of surgical complications. These complications, particularly mucosal tears that could lead to nasal septal perforation, adversely affect patients’ quality of life. ⁹

The findings from this study have significant implications for clinical practice and patient care in nasal surgery. This is apparent when having the majority of patients with thin intraoperative NSMF as shown in the study 76%. The potential for CT scan measurements as a reliable preoperative assessment tool for nasal septal mucoperichondrial flap thickness is highlighted, which could facilitate surgical planning and potentially mitigate the risk of intraoperative complications like mucosal tears. Incorporating CT measurements alongside other known risk factors for septal perforation, such as untreated bilateral corresponding mucosal tears, could enhance the predictability and management of surgical outcomes. ¹ The establishment of a 3.65 mm cutoff for distinguishing thin and thick flaps offers a practical guideline that could be generalized to improve surgical outcomes, Figure 5 shows an example of a thin and thick septum utilizing the cutoff value.

OPEN IN VIEWER

Figure 5.

An example of thin (A) and thick septum (B) on CT scans. CT, computed tomography.

In Huntley et al.’s ¹⁰ research, among 21 patients, 52.4% (11 patients) experienced nasoseptal flap perforations during the harvesting process, with perforation sizes between 3 mm and 15 mm, averaging at 7.3 mm, primarily at the maxillary crest or septal spur. Despite these perforations, no postoperative cerebrospinal fluid leaks were reported. However, the findings’ broader applicability is limited due to the study’s small sample size and study design.

The resolution of CT scans introduces a potential limitation in accurately measuring the septum thickness. While high-resolution scans enhance detail, they may not eliminate inherent variability in measurements due to software limitations in precisely placing ruler borders. This limitation underscores the need for careful interpretation of CT-based thickness assessments and suggests the exploration of more direct or advanced measurement techniques as a future standard for accurate septum thickness evaluation.

In addition, factors like the nasal cycle or the use of decongestants, which might affect the septum thickness, warrant consideration in subsequent studies. ¹¹ Incorporated into our study limitations, we acknowledge that due to the scarcity of prior research in this field, we were unable to estimate an a priori sample size. Consequently, we included all patients within a designated time frame instead.

Further research is warranted to affirm the proposed classification system and to investigate whether individuals with thinner septum are more susceptible to complications such as intraoperative tears, enduring septal perforations during septoplasty, and cerebrospinal fluid leaks in skull base repair utilizing nasoseptal flaps.

In conclusion, this study validates the use of a 3.65 mm threshold on CT scans to categorize nasal septal mucoperichondrial flaps as “thin” or “thick,” aligning well with surgical observations. This standardization could enhance surgical planning, reduce complications, and improve patient outcomes. The research encourages the integration of such measures in preoperative routines and calls for further studies to advance measurement methods for greater precision and clinical relevance.

Supplemental Material