Zero‐Degree Endoscopic Visualization of the Frontal Sinus Predicts Improved Topical Irrigation Delivery

Management of chronic frontal rhinosinusitis is challenging with high rates of treatment failure, exacerbated by limitations of topical irrigation delivery. We hypothesize that intraoperative zero‐degree visualization of the frontal sinus predicts improved postoperative irrigation penetration. Extending a Draf IIa frontal sinusotomy with a limited resection of the middle turbinate axilla‐agger nasi complex can allow zero‐degree endoscopic visualization of the frontal sinus. This study investigates the change in frontal sinus irrigation delivery after standard Draf IIa frontal sinusotomy versus further resection to achieve zero‐degree visualization.


INTRODUCTION
In patients with chronic rhinosinusitis (CRS) refractory to medical management, surgical enlargement of sinus ostia relieves obstruction and improves the delivery of topical irrigation, corticosteroids, and antibiotics, important components of long-term disease control. 1 However, eradication of frontal sinus disease in particular remains one of the most challenging aspects of CRS management. Of patients undergoing primary endoscopic sinus surgery (ESS) with frontal sinusotomy, 7.6% to 8.9% will require a second frontal sinus procedure in the future. [2][3][4] This figure increases substantially (17. .9%) in patients undergoing a revision frontal sinus operation. 2,3 Mucosal disease and retained frontoethmoidal partitions are implicated in the vast majority of frontal sinus treatment failures. 4,5 The narrow frontal sinus ostium makes it susceptible to obstruction and scarring, thus contributing to higher rates of treatment failure. In recent years, topical irrigation delivery access has been increasingly recognized as a primary benefit of ESS. However, frontal sinuses, even in patients who have undergone Draf IIa procedures, receive limited topical delivery of irrigants. 6,7 Poor delivery of irrigation and nasal corticosteroids likely contribute to the higher rates of observed treatment failure in inflammatory sinus disease.
The average dimensions of the frontal sinus ostium are 7.22 × 8.92 mm. 8 The anterior border of the frontal sinus outflow tract is defined by the agger nasi cell. The standard Draf IIa frontal sinusotomy involves opening the posterior wall of the agger nasi, the so-called "uncapping of the egg" technique. 9 Some have studied the additional utility of removing the anterior wall of the agger nasi and the axilla of the middle turbinate in their approach to the frontal sinus for improved visualization of the frontal recess. 10,11 We hypothesize that achieving intraoperative zerodegree visualization of the frontal recess by resecting the middle turbinate axilla-agger nasi complex (MTAN) improves postoperative irrigation penetration. The MTAN is formed by two segments of bone: the frontal process of the maxilla and the nasal process of the frontal bone. The objective of this study is to investigate the change in frontal sinus irrigation following this extension of a standard Draf IIa in a novel cadaver trephination model.

METHODS
Five fresh cadaveric heads were obtained from Science Care (Phoenix, Arizona). The average age of the cadaver specimens was 71.4 years old (range 39-89), consisting of four males and one female. The study was conducted in a surgical skills laboratory at the Columbia University Irving Medical Center and is exempt from Institutional Review Board approval because no live human subjects were affected. Thin cut CT sinus scans were analyzed pre-study for each cadaver to confirm that there was no evidence of prior sinus surgery.

Frontal Sinus Trephination Technique
Frontal sinus trephines were performed in a standardized location 1 cm lateral to the midline at the level of the supraorbital rim. This location has been previously described as a consistent and safe trephine site. 12 The trephine was made using a 5-mm drill bit to accommodate a 4-mm 70-degree endoscope looking inferiorly during endonasal irrigation.

Surgical Technique
Routine endoscopic sinus surgery was performed. This included bilateral uncinectomy, maxillary antrostomy, total ethmoidectomy, sphenoidotomy, and standard Draf IIa frontal sinusotomy, including removal of all frontoethmoidal, suprabulla, and intersinus septal partitions. The anterior wall of the agger nasi was left intact.

Irrigation Technique
A 70 − degree endoscope was inserted into the frontal sinus trephine on one side, obtaining a clear view of the frontal sinus and the frontal recess. The cadaver head was securely positioned  at an angle of 45 degrees head-down, to approximate the conventional head position for saline irrigation. A NeilMed Sinus Rinse bottle (NeilMed Pharmaceuticals, Inc. Santa Rosa, CA) filled with 240 mL of water was irrigated into the ipsilateral naris. The entire bottle was emptied in two squeezes of the bottle. Video recording captured footage of the frontal sinus during irrigation. The same procedure was performed on the contralateral side. To maintain uniformity, one author performed both preand post-irrigations on all cadavers.

Middle Turbinate Axilla-Agger Nasi Complex Resection
Using a 2-mm Kerrison bone rongeur, the axilla of the middle turbinate was resected along with the anterior face of the agger nasi bilaterally to achieve zero-degree endoscopic visualization of the frontal sinus ( Figs. 1 and 2). Zero-degree endoscopic visualization of the frontal sinus was defined as obtaining a clear view of the posterior table of the sinus. The middle turbinate itself remained attached in situ. This procedure was performed on each cadaver by the senior author to maintain uniformity. The irrigation procedure was again performed and recorded on each side in the same fashion.

Endoscopic Evaluation
Each video recording was edited and cropped to ensure anonymity. The recordings were distributed to three blinded reviewers, two attending rhinologists and one rhinology fellow. The videos were reviewed in random order. Scores were assigned separately by each reviewer according to the following scale: 0 = irrigation restricted to nasal cavity; 1 = irrigation reaches frontal recess; 2 = irrigation reaches frontal sinus proper; 3 = irrigation fills entire frontal sinus (Table I). Representative still shots corresponding to each fill level denoted by the scale are seen in Figure 3.

Statistical Analysis
Data was analyzed using STATA version 15.1 (STATA, College Station, TX). Median scores were calculated for each endoscopic recording. The two-tailed Wilcoxon Matched-Pair Signed-Rank test was used to compare groups given the nonparametric nature of the data. Alpha < 0.01 was utilized. Interrater reliability was calculated using Cronbach's alpha to ensure consistency amongst the blinded reviewers. Cronbach's alpha is calculated on a scale of 0-1, with higher ratios signifying greater consistency. 13

RESULTS
Five cadavers were utilized and the above procedure was performed on all 10 frontal sinuses. Irrigant penetration following standard Draf IIa frontal sinusotomy improved after the axilla-agger nasi complex was resected to achieve zero-degree endoscopic visualization of the frontal sinus. The median endoscopic score for irrigant penetration following standard Draf IIa procedure was 2 (interquartile range [IQR]: 1-2). The median endoscopic score for irrigant penetration after additional dissection of the MTAN complex increased to 3 (IQR: 2-3; P < .01; Table II). Inter-rater reliability of the three reviewers was calculated at 0.966, as measured by Cronbach's alpha, signifying excellent consistency between reviewers.

DISCUSSION
The first important revelation of this research is the relatively limited penetration of irrigation even after a standard frontal sinusotomy with a Draf IIa technique to open the frontal recess. This study demonstrates a statistically significant improvement in the volume of irrigation that penetrated the frontal sinus after performing a limited resection of the MTAN complex to achieve zerodegree visualization of the frontal sinus. This finding suggests that these structures directly obstruct the trajectory of irrigant from the nares to the frontal recess in a 45-degree head-down model, which approximates a patient leaning forward over a sink. Improved delivery of topical therapy to the frontal sinus may result in a lower rate of treatment failures.
Three studies have investigated the effects of various frontal sinus procedures on frontal sinus irrigation. Barham et al. compared Draf III to Draf IIa frontal sinusotomy and noted dramatic improvement in irrigation in the Draf III cohort. 6 Using a similar grading scale to our study, Gantz et al. demonstrated moderate irrigant penetration after either balloon sinuplasty or Draf IIa, as compared to no intervention. 7 Lastly, Kidwai et al. hypothesized that removal of the middle turbinate would increase irrigation. 14 This technique improved overall penetration, but this finding was not statistically significant when evaluating the frontal sinuses alone. In contrast, we hypothesized that the MTAN complex is responsible for obstructing irrigant penetration to the frontal sinus. The advantage of this technique is that it allows preservation of the middle turbinate.
The vector of a zero-degree endoscope pointed toward the fontal recess mirrors the trajectory of irrigant reaching the frontal recess (Fig. 4). This relationship allows the surgeon to utilize his or her visualization as a proxy for postoperative frontal sinus topical irrigation access. In most patients, the described technique will allow only visualization of the posterior table, rather than the entire lumen of the sinus. However, even this degree of visualization corresponded to improved irrigation penetration.
Achieving zero-degree visualization of the frontal sinus may also facilitate postoperative visualization in the office. By lowering the angle of a zero-degree telescope down to the nasal sill, rather than up to the soft tissue triangle of the nose, the frontal sinus posterior table can often be visualized adequately in the awake patient (Fig. 5).
In this study, we also describe a new model to evaluate frontal sinus irrigation. Similar to the study reported by Barham et al. we recorded irrigation via a frontal sinus trephine. 6 However, we performed frontal sinus trephines 1 cm lateral to midline at the supraorbital rim because this is the safest location to make a trephine that will reliably be lateral to the intersinus septum and remain within the sinus. 12 In our experience, the use of a 70-degree endoscope provided the best visualization.
This study is limited by its small sample size and the cadaveric nature of the model. While many hypothesize that insufficient delivery of topical irrigation contributes to treatment failure, well-designed studies are necessary to evaluate rates of failed frontal sinusotomy with MTAN complex resection.

CONCLUSION
This study demonstrates improved penetration of nasal cavity irrigation to the frontal sinuses following removal of the MTAN complex to achieve intraoperative zero-degree visualization of the frontal sinus. This technique achieved significantly improved irrigation penetration compared to a standard Draf IIa frontal sinusotomy in a novel cadaver trephine model. This finding may guide the extent of intraoperative surgical dissection as a proxy for achieving optimal postoperative topical irrigation access.