Available online at www.sciencedirect.

com

American Journal of OtolaryngologyHead and Neck Medicine and Surgery 31 (2010) 175 180 www.elsevier.com/locate/amjoto

Operative findings in the frontal recess at time of revision surgery

Kristen J. Otto, MD, John M. DelGaudio, MD
Department of Otolaryngology-Head and Neck Surgery, The Emory Clinic, Atlanta, GA, USA Received 18 September 2008

Abstract

Objective: Endoscopic sinus surgery is the gold standard for the treatment of medically refractory chronic rhinosinusitis. There is, however, a population of patients for whom persistent disease is a problem. Of all the sinuses, the frontal sinus is the most likely to have recurrent obstruction. We evaluated the findings causing frontal recess obstruction at the time of revision surgery. Study design and setting: A retrospective review was performed in a tertiary care academic otolaryngology department. Results: Findings obstructing the frontal recess at the time of revision sinus surgery were reviewed. Two hundred eighty-nine frontal sinuses were included. Seven findings were identified: mucosal disease (67%), retained ethmoid cells (53%), lateralized middle turbinates (30%), retained agger nasi cells (13%), scar (12%), retained frontal cells (8%), and neoosteogenesis (7%). Most frontal recesses had multiple etiologies for failure listed above, with an average of 1.6. Conclusions: Multiple findings can be identified that contribute to frontal recess obstruction requiring revision sinus surgery. A comprehensive approach to address all factors is necessary to prevent surgical failure among patients presenting for endoscopic frontal sinus surgery. 2010 Elsevier Inc. All rights reserved.

1. Introduction Endoscopic sinus surgery has proven to be largely successful for most patients at preventing symptom recurrence and leading to a disease-free state. Success rates as high as 97.5% have been reported [1]. The frontal sinuses, however, still represent a topic of controversy and debate when it comes to how to effectively deal with them surgically, how to manage them medically, and how to address the surgical failures. The purposes of this study are to document findings in the frontal recess in patients requiring revision endoscopic sinus procedures and to discuss how these findings ultimately led to surgical failure. 2. Patients and methods All patients requiring revision endoscopic sinus procedures that included the frontal sinus between May 1997 and
The authors have no relevant financial interest in this article. Corresponding author. Department of Otolaryngology-Head and Neck Surgery, The Emory Clinic, 1365A Clifton Road NE, Suite 2100, Atlanta, GA 30322, USA. Tel.: +1 404 778 3382; fax: +1 404 778 4295. E-mail address: jdelgau@emory.edu (J.M. DelGaudio).

October 2003 were reviewed. Patients were evaluated both by in-office endoscopic examination under topical anesthesia and noncontrast computed tomography (CT) scan with multiplanar reformatting. Patients were offered revision endoscopic surgery based on recurrence of symptoms and by objective endoscopic and CT findings of sinus disease, refractory to medical therapy. Preoperative medical management included combinations of a 30-day course of antibiotics, a 2- to 3-week course of systemic corticosteroid taper, as well as topically administered steroid drops or sprays as deemed necessary by the senior author. Patients included in the study were those patients whose frontal recesses were addressed during the revision operation. Institutional review board approval was obtained before patient chart review. A total of 149 cases were analyzed retrospectively for operative findings contributing to disease in the frontal sinuses. Data on sex, associated conditions, and associated disease states (ie, Samter's triad, cystic fibrosis, allergic fungal sinusitis) were also collected and entered into a database. All revision operations were performed by the senior author (J. M. D.). Revision operations were all performed under general anesthesia using image guidance. Procedures performed were tailored to each patient's underlying frontal recess anatomy and

0196-0709/$ see front matter 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.amjoto.2008.12.006

176

K.J. Otto, J.M. DelGaudio / American Journal of OtolaryngologyHead and Neck Medicine and Surgery 31 (2010) 175180

findings and included release of scarring and adhesions, opening of residual anterior ethmoid air cells, polypectomy, and takedown of neoosteogenic bone within the frontal recess. Concurrent maxillary antrostomy, ethmoidectomy, and sphenoid sinusotomy were performed when necessary. A total of 298 frontal sinuses from 149 cases of revision endoscopic sinus surgery were reviewed. Two sides were excluded from the study because they had not been previously addressed and were thus considered primary cases. Seven more sides were excluded because they were felt to be disease-free both by history and physical examination. These sides did not require attention at the time of revision surgery. The study group was defined as the remaining 289 frontal sinuses requiring revision endoscopic surgery for symptoms and physical and radiologic findings of frontal sinus disease. 3. Results The study group of 149 cases included 127 different patients, 61 (48%) male and 66 (52%) female. There were 12 patients requiring multiple revision procedures during the study dates. The number of revision procedures within this group ranged from 2 to 7. There were 7 major factors identified as important frontal recess findings after previous endoscopic sinus surgery. In most cases, multiple findings were evident in each frontal recess (Table 1). One of the most common findings was inflammatory mucosal disease, being evident in 193 (67%) revision frontal recesses. This includes polyps and nonpolypoid inflammatory edema obstructing the frontal recess and/or frontal sinus. The other most common finding was retained sinus cells or septations obstructing the frontal recess. This was further
Table 1 Graph shows distribution of pathologies found in frontal recesses in this study Fig. 1. A sagittal CT scan of a patient with a retained agger nasi cell and a retained suprabullar cell causing frontal recess obstruction after previous endoscopic sinus surgery. Dot is on obstructed frontal outflow tract.

Retained ethmoid cells are shown in same vertical bar that is subdivided into the various ethmoid derived cells. Patients had an average of 1.6 findings per frontal recess.

broken down into specific sinus cells. A total of 213 (74%) frontal sinuses appeared to be obstructed by retained ethmoid cells. Of these, 37 (13%) were agger nasi cells, and 153 (53%) were other anterior ethmoid cells, such as the ethmoid bullae, suprabullar cells, frontal bullar cells, or supraorbital ethmoid cells. In addition, retained frontal cells were found in 23 (8%) frontal recesses (Figs. 1, 2, and 3). The next most frequent intraoperative finding was a middle turbinate that had scarred laterally as a result of previous surgery, causing obstruction of the frontal sinus outflow tract. Lateralized middle turbinates were found in 88 (30%) frontal recesses (Fig. 4). Scar in the frontal recess separate from the lateralized middle turbinate was found in 34 (12%) frontal recesses. Neoosteogenesis, representing new bone formation or osteitis, was present in 19 (7%) frontal recesses (Fig. 5). In 12 (4%) cases, the frontal recesses were found to have disease by preoperative evaluation, but on exploration, no pathology was found and an open frontal sinus ostium was noted. This is likely due to resolution of mucosal inflammation by preoperative steroid use Each frontal sinus analyzed was noted to have between 1 and 4 (mean of 1.6) of the above findings. In each case, the findings were thought to contribute to overall sinus failure. In addition, some of the operative findings tended to occur together. Of the 37 frontal recesses that were noted to have a retained agger nasi cell, 67.6% also had another retained anterior ethmoid cell. Of the frontal recesses found to have retained anterior ethmoid cells, one third were also noted to have a lateralized middle turbinate. Of the sinuses found to have scar in the frontal recess, almost half (47%) also had a lateralized middle turbinate as well. Of the frontal recesses with neoosteogenesis, scar tissue was also found in 36.8% of these cases. In the group of patients requiring multiple revisions, there were 3 patients with allergic fungal sinusitis, 1 patient with

K.J. Otto, J.M. DelGaudio / American Journal of OtolaryngologyHead and Neck Medicine and Surgery 31 (2010) 175180

177

Fig. 2. Triplanar images of a retained agger nasi cell causing frontal recess obstruction: sagittal (A), axial (B), and coronal (C) images. Dot is on superior wall (cap) of agger nasi cell.

Samter's triad, 1 patient with cystic fibrosis, 1 patient with Wegener granulomatosis, 1 patient with sarcoidosis, and 5 with chronic rhinosinusitis. One patient with chronic rhinosinusitis requiring the most (7) revision procedures had had a previous modified Lothrop procedure for recalcitrant frontal sinus disease. This patient was noted on endoscopic exploration to have a residual frontal cell initially and scarring in the frontal ostium and frontal recess at subsequent procedures. In the entire study group, concurrent diagnoses were Samter's triad (7 patients), allergic fungal sinusitis (14 patients), cystic fibrosis (2 patients), and chronic rhinosinusitis (92 patients). Seventeen patients were operated on for the finding of a mucocele, either of the frontal

sinus or of the anterior ethmoids. One patient was noted to have trauma as the major cause of his chronic frontal sinusitis, having previously undergone anterior table frontal sinus fracture repair.

4. Discussion Despite the fact that endoscopic sinus surgery is the most commonly performed surgical procedure for the treatment of chronic rhinosinusitis and despite recent significant advancements in the imaging and instrumentation available to sinus surgeons, the frontal sinuses still represent a management

178

K.J. Otto, J.M. DelGaudio / American Journal of OtolaryngologyHead and Neck Medicine and Surgery 31 (2010) 175180

failure, he attributed 31% of his 52 failures solely to retained ethmoid air cells. In his study, he reviewed 398 primary cases of endoscopic sinus surgery, of which 52 cases ended in revision. His series excluded patients with severe polyposis and systemic illnesses such as immunodeficiency or sarcoidosis. In evaluating the cause for failure in each of his cases, he assigned only 1 major factor to each case, and retained ethmoid air cells were the most common finding. Because multiple factors may contribute to persistent frontal sinus disease after endoscopic sinus surgery, we chose to include all contributors to frontal recess obstruction identified in each case. In addition to the above-mentioned ethmoid cells, 23 (8%) frontal recesses in our study were found to contain a frontal cell. Preoperative planning with dedicated sinus
Fig. 3. A coronal CT scan of a retained type III frontal cell after endoscopic sinus surgery.

challenge. The frontal sinus ostium and frontal recess in particular are prone to stenosis after surgical manipulation because of difficult anatomy and the limited dimensions of the frontal recess. The predilection of the frontal sinus outflow tract to recurrent mucosal disease has also been demonstrated [2]. In our study group, two thirds (193/289) of sinuses were noted to have mucosal edema or polyposis obstructing the frontal recess at the time of endoscopic exploration for revision surgery. For patients with severe polyposis or inflammation of the paranasal sinus mucosa after a failed endoscopic surgery, a regimen of thoroughly revisiting the preoperative history, nasal endoscopy, CT scanning, and anti-inflammatory medical therapy is indicated. Mucosal inflammatory disease in the frontal recess, in the absence of anatomic reasons for obstruction, should be considered a medical problem rather than a surgical one [3]. If there are no retained cells, scar, or neoosteogenesis identifiable in the frontal recess, only mucosal inflammation, many cases will respond to medical treatment alone. In fact, even if these patients undergo surgery for the inflammatory mucosa, aggressive medical treatment is required to control the inflammation postoperatively. Polyposis in the general population is a common finding, estimated to affect between 1% and 4% [4]. In all, because polyposis is the final common pathway for a number of host physiologic processes, both allergic and nonallergic [4], a complete workup must be performed to achieve an underlying diagnosis so that treatment may be correctly and individually tailored for maximal patient benefit and longterm control. This may include screening for Samter's triad (salicylate sensitivity), bacterial and fungal cultures, and allergy testing. Nearly three quarters of the sinuses in our study group (213/289) were also found to have retained ethmoid cells obstructing frontal sinus outflow. (Figs. 1, 2, and 3). In a study by Ramadan [1] looking at all sinuses and reasons for

Fig. 4. Evidence of a lateralized middle turbinate causing frontal recess obstruction after endoscopic sinus surgery. (A) Frontal sinus opacification is evident on the CT with the right middle turbinate adherent to the lamina. (B) Endoscopic view of a left middle turbinate scarred to the lateral nasal wall (an adhesion between the turbinate and septum is present as well).

K.J. Otto, J.M. DelGaudio / American Journal of OtolaryngologyHead and Neck Medicine and Surgery 31 (2010) 175180

179

Fig. 5. A sagittal CT scan showing neoosteogenesis of the frontal recess contributing to obstruction of the frontal sinus. Crosshairs are on the affected bone. Note the retained ethmoid bulla behind the new bone.

imaging assists with identifying frontal cells within the sinus or the frontal recess. Both parasagittal and coronal views are helpful in this regard. When frontal cells are encountered and opened during endoscopic procedures, they are often mistaken for the frontal sinus. Adequate ventilation of the sinus in these cases is oftentimes not achieved. Frontal cells in the general population are common, and that makes awareness of their presence important for the rhinologist. Meyer et al [5] reported that of 768 coronal sinus CT scans reviewed, 20.4% were noted to have frontal cells. DelGaudio et al [6] found that 33% of patients undergoing endoscopic sinus surgery have frontal cells. The use of computer-guided surgery greatly assists with identification of such cells within the frontal sinus and recess and should especially be considered in cases of recurrent frontal sinus disease [7]. Image guidance greatly assists the sinus surgeon in navigating the tight confines of the frontal recess and allows identification of landmarks that would otherwise be missed and potentially lead to surgical failure [3]. These findings stress the importance of a complete ethmoidectomy, including adequate removal of agger nasi cells, in ensuring frontal sinus ventilation. Careful preoperative planning to assist with complete surgery, including reviewing a CT scan (performed after maximal medical therapy) for agger nasi cells, supraorbital ethmoid cells, frontal cells, and other ethmoid cell variations, is essential. The use of parasagittal views provides the best preoperative imaging modality for surgery involving the frontal recess. Complete uncapping of the agger nasi cell is necessary for complete clearing of the frontal recess [3] (Fig. 2). Approximately one third (88/289) of the frontal sinuses in our study were affected by a lateralized middle turbinate.

This could be the result of overaggressive resection and/or weakening the support for the middle turbinate. Mucosal stripping leading to exposed bone and local inflammation causes synechiae formation between the middle turbinate and lateral nasal wall, leading to a blockage of the frontal recess. This problem can be exacerbated by incomplete uncinectomy where the uncinate remnant comes in contact with the middle turbinate [8]. Ramadan notes that most of his patients with residual ethmoid air cells were also found to have a lateralized middle turbinate that was adherent to the residual cells [1]. In our study, 32.7% of sinuses with retained ethmoid cells also had a lateralized middle turbinate. To ensure a stable medially positioned middle turbinate, the sinus surgeon should attempt careful preservation of the horizontal basal lamella. It is also important to ensure that the anterior and superior attachments of the middle turbinate are left in place and not traumatized [9]. In cases where middle turbinate destabilization is inevitable, various techniques have been described to medialize the middle turbinate. Sometimes, limited resection of the anterior aspect of a flaccid middle turbinate is necessary to prevent lateral scarring [9] (Fig. 4). Scar in the frontal recess was a finding in 34 (12%) of 289 sinuses in our series. Scar formation can occur as the result of mucosal trauma, exposed bone, persistent inflammatory disease, or retained air cells. Because of the narrow confines of the frontal recess, any unnecessary manipulation can lead to development of adhesions and stenosis. Scarring in the frontal recess was thought to be the major cause of frontal sinus obstruction in 25% of revision cases studied by Ramadan [1]. A study by Bhattacharyya [10] addressed whether frontal sinusotomy at the time of primary endoscopic sinus surgery leads to improved outcomes. He reported 35 frontal sinuses without disease at initial surgery, of which 7 eventually developed disease requiring revision. Of 49 frontal sinuses with disease at the time of primary endoscopic sinus surgery, 21 underwent sinusotomy and 19 developed recurrent disease. Of the 28 that did not undergo initial sinusotomy, 23 eventually presented with recurrent or persistent disease. His data suggest that although no correlation exists between sinusotomy and the development of frontal sinus disease, performing a frontal sinusotomy for frontal disease at the time of initial surgery does not result in less frequent disease at the time of revision. Overall, he recommends leaving the frontal sinus alone initially. To prevent excessive scar formation and lessen the impact of the inflammation induced by even minimal work in the frontal recess, an aggressive postoperative regimen of active clot, debris, and mucous removal is the recommendation of most sinus surgeons [8]. We advocate that the frontal recess should be directly addressed either in its entirety, or not at all. The surgeon should decide whether to avoid manipulation of the frontal recess or to completely clear the frontal recess of all cells to allow unimpaired outflow of the frontal sinus. Partial clearing of the frontal recess may cause edema and scar in an area already compromised by its limited dimensions.

180

K.J. Otto, J.M. DelGaudio / American Journal of OtolaryngologyHead and Neck Medicine and Surgery 31 (2010) 175180

Of 298 sinuses in our study, 19 (17%) had frontal recess obstruction by neoosteogenic bone. Neoosteogenesis, as it relates to chronic sinusitis, is inflammationinduced osteoblastic activity in the periosteum resulting in new, sclerotic bone formation [11]. The role of bone involvement in chronic sinusitis is still poorly understood, but inflammatory changes in the bony septations of the sinuses, similar to that in chronic osteomyelitis in long bones, has been demonstrated [12]. These findings are often evident on CT scan and can be confirmed by endoscopic examination (Fig. 5). Osteitic bone associated with overlying diseased mucosa may require aggressive anti-inflammatory and antibiotic therapy, and in persistent cases, revision surgery. Unfortunately, revision surgery with removal of neoosteogenic bone often results in further neoosteogenesis. Overall, the key factor to preventing inflammation and scarring leading to recurrent sinus disease is meticulous preservation of the mucosal integrity within the frontal recess. Mucosal disruption can lead to scar, granulation, and mucociliary dysfunction as the mucosa regenerates in an attempt to cover exposed bone. Although it is possible that the long-standing inflammatory changes within the sinus lead to physiologic outflow obstruction by affecting ciliary function, it is also possible that previous traumatic manipulation caused functional obstruction of the frontal recess. Surgical intervention should strive to minimize mucosal trauma and maintain mucociliary clearance, promoting normal sinus physiology [13]. Postoperative use of topical steroid drops or other preparations in the frontal recess has been shown to be effective in reducing inflammation, granulation, and scarring in the frontal recess in the postoperative period. This can help to maintain frontal recess patency [14]. Because all of our data were reviewed and collected retrospectively, it is difficult to be certain whether each finding represents a cause of subsequent frontal sinus failure. What is important to consider, however, is the fact that the operative findings that appeared to cause obstruction of the frontal recess were directly related to previous surgical manipulation 75% of the time. One of the limitations of this review was the lack of historical data surrounding each of our patients' clinical histories. Because most patients presented as referrals after primary endoscopic procedures had already been performed at outside institutions, information was not available as to each patient's original extent of disease, CT, or endoscopic findings. Also unknown was the goal of the initial operation. Some of these patients likely had attempted frontal sinusotomy for frank frontal sinus disease, whereas others my have developed frontal sinus pathology after more minimal surgery for maxillary or ethmoid disease.

5. Conclusion Frontal sinus disease is present 48% to 63% of all revision sinus cases, suggesting that frontal sinusitis is a significant factor in overall failures [10]. A review of the patients presenting for revision frontal sinus surgery at our institution revealed 7 reproducible findings that contributed to frontal sinus failure. Recurrent mucosal disease, retained ethmoid air cells, and lateralized middle turbinates were the most common findings, whereas missed frontal cells, scarring, and neoosteogenesis were also found to contribute. One of the issues not addressed in this study was the type of operation performed either initially or at the time of revision. This information has the potential to contribute which procedures lead specifically to which causes for failure. Although the best approach to the frontal sinus is still the subject of debate, there is no question that a comprehensive approach including aggressive medical management, meticulous technique with mucosal preservation and minimal mucosal trauma, complete removal of all diseased cells, and thorough postoperative care can lead to improved outcomes and fewer technical failures. References
[1] Ramadan HH. Surgical causes of failure in endoscopic sinus surgery. Laryngoscope 1999;109:27-9. [2] Smith TL, Rhee JS, Loehrl TA. Surgical management of frontal sinusitis. Curr Opin Otolaryngol Head Neck Surg 2001;9:42-7. [3] Sonnenburg RE, Senior BA. Revision endoscopic frontal sinus surgery. Curr Opin Otolaryngol Head Neck Surg 2004;12:49-52. [4] Chiu AG, Kennedy DW. Disadvantages of minimal techniques for surgical management of chronic rhinosinusitis. Curr Opin Otolaryngol Head Neck Surg 2004;12:38-42. [5] Meyer TK, Kocak M, Smith MM, et al. Coronal computed tomography analysis of frontal cells. Am J Rhinol 2003;17:163-8. [6] DelGaudio JM, Beningfield A, Hudgins PA, et al. Multiplanar CT analysis of frontal recess cells: impact on frontal isthmus dimensions and the presence of frontal sinusitis. Arch Otolaryngol Head Neck Surg 2005;131:230-5. [7] Loehrl TA, Toohill RJ, Smith TL. Use of computer-aided surgery for frontal sinus ventilation. Laryngoscope 2000;110:1962-7. [8] Chow JM. Technical reasons for endoscopic sinus surgery failures. Curr Opin Otolaryngol Head Neck Surg 2002;10:33-5. [9] Bolger WE, Kuhn FA, Kennedy DW. Middle turbinate stabilization after functional endoscopic sinus surgery: the controlled synechiae technique. Laryngoscope 1999;109:1852-3. [10] Bhattacharyya N. Computed tomographic staging and the fate of the dependent sinuses in revision endoscopic sinus surgery. Arch Otolaryngol Head Neck Surg 1999;125:994-9. [11] Kocak M, Smith TL, Smith MM. Bone involvement in chronic rhinosinusitis. Curr Opin Otolaryngol Head Neck Surg 2002;10:49-52. [12] Perloff JR, Gannon FH, Bolger WE, et al. Bone involvement in sinusitis: an apparent pathway for the spread of disease. Laryngoscope 2000;110:2095-9. [13] Catalano PJ. Minimally invasive sinus technique: what is it? Should we consider it? Curr Opin Otolaryngol Head Neck Surg 2004;12:34-7. [14] DelGaudio JM, Wise SK. Topical steroid drops for the treatment of sinus ostia stenosis. Am J Rhinol 2006;20:563-7.