The success of 6-month stenting in endonasal frontal sinus surgery.
The ultimate success or failure of frontal sinus surgical procedures, whether they be endonasal or external, is determined essentially by the rate of restenosis of the frontal sinus outflow tract or neo-ostium postoperatively. Long-term stenting for a period of several months significantly reduces the rate of restenosis, particularly in difficult cases. We retrospectively reviewed the cases of 12 patients who received 21 frontal nasal stents, which were left in place for 6 months. Based on outcomes measures that included endoscopy or radiologic findings and patients' self-evaluations, we conclude that frontal nasal stents that are left in place for 6 months are more effective than stents that are removed earlier. We recommend that this type of management be considered in difficult revision cases and before performing an external operation.
Endonasal treatment of frontal sinus inflammatory disease has become increasingly established during the past few years. [1-14] The ultimate success or failure of frontal sinus surgical procedures, whether they be endonasal or external, is determined essentially by the rate of restenosis of the frontal sinus outflow tract (FSOT) or the neoostium postoperatively. The spectrum of endonasal procedures includes (1) exposure of the frontal recess with preservation of the natural ostium, (2) resection of the frontal sinus floor between the lamina papyracea and the nasal septum, and (3) median drainage with removal of the frontal sinus floor on both sides and removal of adjacent parts of the interfrontal and nasal septum, classified as Draf types I, II, and III (table 1). [2,4,13,15,16]
In about one-third of cases when the frontal sinus is opened intraoperatively, the passage of a probe will not be possible after healing.  This proportion can be reduced by performing an intensive intraoperative dissection at the frontal sinus ostium [9,13] or by creating a Draf type III drainage.  In both cases, the more extensive procedure is reported to carry a lower risk of restenosis and confer a greater likelihood of mucosal healing. However, in the case of type III drainage in particular, the surgical trauma is not insignificant. The more extensive removal of bony structures with a drill leads to a more marked postoperative inflammatory response, and some healthy structures (e.g., parts of the nasal septum, the interfrontal septum, the anterior attachment of the middle turbinate, and parts of the frontal process) must be sacrificed to achieve wide drainage. Despite the creation of a very wide opening intraoperatively, it cannot be guaranteed that the opening will not close completely during th e succeeding weeks or months.
Endonasal frontal sinus surgery fails when the specific pathology cannot be reached or when a long-lasting stabile drainage cannot be established. In these cases, an external approach is necessary. The osteoplastic flap procedure with fat obliteration has been hailed as the gold standard of definitive frontal sinus procedures. [17-19]
To minimize the surgical trauma and scarring and to avoid larger operations, particularly the external approach, stenting of the FSOT and neo-ostium should be considered. The decisive point is that the duration of stenting placement must be long enough so that wound healing and scar remodeling are nearly complete. It has been shown that long-term stenting for a period of 6 months significantly reduces the rate of restenosis of the frontal sinus neo-ostium. 
In this article, we describe our experience with stenting of the FSOT and neo-ostium in difficult cases in patients who had had recurrent polypoid disease and major scarring at the FSOT.
Patients and methods
We retrospectively evaluated 12 patients who had received a frontal sinus stent that had been placed to keep the FSOT open (table 2). Eight of these patients had had recurrent chronic polypoid sinusitis with major involvement of the frontal sinus. Their FSOTs had been closed by scar tissue. The decision to introduce a frontal sinus stent into these patients was based on the intraoperative observance of major areas of denuded bone at the completion of the operation. The indication for revisional endonasal frontal sinus surgery was based on therapy-resistant headache and corresponding signs on computed tomography (CT).
Two other patients had had an osteoma of the frontal sinus. After the osteoma was completely removed endonasally, the mucosa of the FSOT was also removed, and denuded bone was seen on all sides. The remaining two patients had had mucopyoceles after previous sinus surgery for major scarring of the FSOT.
None of these patients wished to have an external operation, and all agreed to undergo an endonasal revision and long-term stenting of 6 months.
Type of procedure.
With a microscope and endoscope, a wide opening on one or both sides of the frontal sinus was made to establish a Draf type II or III drainage.  Depending on the individual situation, three types of stents were inserted.
One stent was a Rains self-retaining silicone tube with a compressible basket at the frontal sinus end. This stent was used for patients with unilateral disease (figure 1).
The second stent was a silicone tube of 5 mm in diameter that was positioned in a U-shaped manner into both FSOTs (figure 2). Insertion was made possible by creating a small perforation in the upper part of the nasal septum. This type of procedure was used for patients who had bilateral disease in order to achieve optimal positioning of the stent and to avoid crusting on the end of the nasal tube inside the middle meatus.
The third type of stent was a self-retaining H-shaped silicone tube, which was used for procedures that involved a type III revision (figure 3).
Postoperative management and follow up. In all cases, the nose was packed for 5 to 7 days with rubber finger stalls. Postoperative treatment included a 6-month course of topical steroids and self-administered nasal irrigation with saline solution delivered by a nasal douche three times daily. There was no routine administration of antibiotics or systemic corticoids.
Endoscopic followup examinations with a flexible endoscope were performed after removal of the stent and at the final followup visit. The frontal sinus opening was determined to be either (1) open so that the frontal sinus mucosa itself could be seen, (2) open but narrow so that the frontal sinus mucosa could not be assessed, (3) not visible because of the presence of polyps or scar tissue in the ethmoid, or (4) closed because of scarring in the FSOT. Additionally, all patients were asked to provide a subjective judgment of their own outcome according to four categories: (1) free of complaints or major improvement, (2) moderate improvement, (3) mild or no improvement, or (4) worse. Patients whose frontal sinus could not be evaluated endoscopically were advised to undergo CT or magnetic resonance imaging (MRI) so that aeration of the frontal sinus could be evaluated.
The goal of the stenting procedure was to prevent the closure of the FSOT by scar tissue. We considered this goal to have been achieved if the frontal sinus could be viewed endoscopically as open or if the frontal sinus was shown by CT or MRI to be aerated.
Altogether, the 12 patients received 21 stents (table 3). Seven sinuses received a Rains frontal sinus stent (three patients unilateral, two bilateral), 10 sinuses received a U-shaped stent (five patients bilateral), and four received an H-shaped stent (two patients bilateral). All stents were removed after 6 months.
After a followup of 10 to 36 months (mean: 19.4), 10 of the 12 patients reported that they were either free of complaints or that they had experienced major improvement (table 4). The remaining two patients judged their improvement to be moderate.
The two patients who had had an osteoma with otherwise healthy mucosa retained their normal mucosa and did not experience any polyp development. Both of their FSOTs were wide open.
All eight patients who had had polypoid sinusitis preoperatively experienced a recurrence of polyps during followup. Three of these patients experienced some superimposition of edema as a result of acute viral rhinosinusitis.
Two of the three type III drainages were not visible on endoscopy because of polyps in the ethmoid (table 4). One of these two patients agreed to an MRI examination, which showed that the frontal sinus was aerated and that the type III drainage was open.
Of the remaining 15 FSOTs (type II drainage), nine were not visi' le on endoscopy because of the presence of ethmoidal polyps. Five were later examined by CT or MRI, and four were not. All five FSOTs that underwent followup radiologic evaluation were found to be ventilated (figure 4). The other six FSOTs were open on endoscopy; edema or small polyps were visible in four, and the frontal sinus was normal in two (figure 5).
The small perforation in the upper part of the nasal septum that was made during the U-shaped stent procedure remained in all five of these patients, but it caused no symptoms.
All in all, surgery and stenting could not prevent the recurrence of local polypoid sinusitis. However, it did prevent functional, relevant stenosis of the FSOT by scar tissue in 15 of the 21 sinuses. For the remaining six sinuses, endoscopic evaluation was hindered by ethmoidal polyps and CT or MRI examination was not available. Therefore, a definitive assessment of these patients was not possible. Nevertheless, they were all free of complaints.
Our study shows that long-term stenting of the FSOT was able to prevent restenosis by scar tissue in a large number of very difficult revision cases. Stenting could not prevent a recurrence of the local regrowth of polyps, but it did accomplish at least partial drainage and ventilation. This led to a major improvement of patients' symptoms and helped them avoid major external surgery.
The reduction in postoperative synechiae and stenosis is regarded as the decisive advantage of stents. According to a literature review by Shikani, scarring occurs in 6 to 15% of cases, and the blockage of the middle meatus leads to a recurrence of sinusitis in up to 7%.  Toffel described his experience in treating more than 3,000 patients.  He inserted a special silicone and Merocel pack for 5 to 7 days under cephalosporin antibiotic cover following combined endoscopic paranasal sinus surgery and septoplasty. A second Merocel sponge was placed in the inferior nasal cavity for gentle compression of the septum, and a ventilation tube was also inserted. Of the 1,363 patients he followed, only 41 (3%) developed synechiae between the middle turbinate and lateral wall 1 to 3 years postoperatively. Revision surgery was necessary in only 15 cases.
With regard to the frontal sinuses, Weber et al reported that 33 of 96 Draf type II drainages (34%) and 16 of 86 Draf type III drainages (19%) were not endoscopically patent because of scarring.  In another report by Weber et al of a subgroup of patients with no frontal sinus stenting, almost half (47.6%) of the frontal sinuses evaluated were not patent on endoscopy because of scar formation. 
Stenting of the frontal sinus via an eyebrow incision was described by Amble et al,  Neel et al, [24,25] and Stammberger.  Schaefer and Close used a silicone catheter for 6 weeks in endonasal frontal sinus surgery.  They reported that 32 of 36 patients (89%) experienced an overall improvement or resolution of symptoms in an average of 16.4 months postoperatively. They did not report any data on endoscopic findings. Hoyt described the endonasal insertion into the frontal sinus of a ventilation tube that was anchored to the anterior nasal septum with a Vicryl suture.  The indications for this procedure were acute or chronic frontal sinusitis in patients who had undergone at least one previous operation and who had had marked frontal sinus pathology on CT. Thirty-two tubes were inserted in 21 patients and left in place for an average of 8.3 weeks. After an unspecified but short followup, the failure rate was 9.5%.
In a prospective study by Weber et al, 80% of the frontal sinus neo-ostia were endoscopically open 12 to 16 months after long-term stenting, compared with only 33% of frontal sinuses that were not stented.  Occlusion by scar tissue was seen in 6.7 and 48% of the two groups, respectively. Wigand and Hosemann found that 3.5 years after sphenoethmoidectomy and enlargement of the frontal infundibulum with a diamond burr (n = 126 sides), the frontal sinus was endoscopically accessible in 40%; probing was possible in 30% and not possible in 30%.  Hosemann et al found an endoscopically accessible frontal sinus in 22% of 154 operations an average of 13 months following endoscopic extended frontal sinus drainage; probing was possible in another 59%.  On the basis of precise intra- and postoperative measurements, Hosemann et al found a decrease in the average minimum diameter from 5.6 mm intraoperatively to 3.5 mm postoperatively. Below an intraoperative minimum diameter of 5 mm, the proportion of ostia tha t became obstructed postoperatively rose sharply, from 16 to more than 30%. As a process of the physiologic mechanisms that are involved in both postoperative wound healing in general [28,29] and wound healing after paranasal sinus surgery, [30-32] resenosis of a surgically enlarged opening of the frontal sinus can take place in three partly related ways.
First, restenosis can occur as a result of a persistent obstruction of the opening by blood and fibrin during the immediate postoperative phase. During the proliferative phase beginning on postoperative day 2 or 3, fibroblasts migrate into the fibrin mesh and form granulation tissue. Collagen is deposited and the opening can become occluded by scar tissue.
Second, the marked swelling that begins during postoperative week 3 [31,32] can lead to the formation of contact zones between adjacent or opposite parts of the frontal sinus opening. As a result of either a primary absence of epithelialization or secondary epithelial damage caused by pressure-induced maceration or inflammatory cells, contact between areas of granulation tissue can lead to the formation of tissue bridges. Here, too, the final stage is occlusion by scar tissue.
Third, stenosis can occur as a result of the realignment of collagen fibers during the remodeling phase that begins during postoperative week 3. In this case, the fibers form "scar leaflets" on concave surfaces because the distance can be shortened only by making use of the free lumen. This can lead to a concentric narrowing of ring-shaped openings.
The insertion of a stent allows epithelialization to take place along the surface of the device. The presence of a stent provides the subepithelial scar layer a chance to stabilize. To be effective, a stent must therefore remain in situ for several months. Six months appears to be appropriate and sufficient. Good results have been achieved in tracheal surgery with the insertion of Montgomery tubes, which are also left in place for several months. Restenosis is considerably less likely to occur when a stent is left in place for 6 months than when it is removed after only 2 or 3 weeks.
Certainly, there are some problems with the use of frontal sinus stents. If the stent is too short, granulations can overgrow it and embed it in scar tissue. If the stent is too long, a persistent crusting of the nasal end will occur, which can disturb the patient and can cause an unpleasant smell. Aspiration can also occur, although adequate fixation of the stent should prevent this.
On the other hand, the self-anchoring design of some modem stents obviates the risk of dislocation. The transseptal insertion of a U-shaped stent is an option for special revision cases in patients who have large wound areas inside the ethmoid cavity and a frontal sinus that needs to be splinted.
From the Department of Otorhinolaryngology, Otto-von Guericke-University of Magdeburg, Germany (Dr. Weber); the Department of Otorhinolaryngology--Head, Neck, and Facial Plastic Surgery, Fulda Hospital, Academic Teaching Hospital of the University of Marburg, Fulda, Germany (Dr. Weber, Dr. Mai, and Dr. Draf); the Department of Otorhinolaryngology, University of Regensburg, Regensburg, Germany (Dr. Hosemann); and the Department of Otolaryngology--Head and Neck Surgery, University of Southern California School of Medicine, Los Angeles (Dr. Toffel).
Reprint requests: Prof. Dr. Rainer Weber, MD, Department of Otorhinolaryngology, Otto-von Gucricke-University of Magdeburg, Leipziger Strasse 44, D-39120 Magdeburg, Germany.
(1.) Close LG, Lee NK, Leach JL, Manning SC. Endoscopic resection of the intranasal frontal sinus floor. Ann Otol Rhinol Laryngol 1994; 103 :952-8.
(2.) Draf W. Endonasal micro-endoscopic frontal sinus surgery, the Fulda concept. Op Tech Otolaryngol Head Neck Surg 1991 ;2:234- 40.
(3.) Draf W, Weber R. Endonasal micro-endoscopic pansinusoperation in chronic sinusitis. I. Indications and operation technique. Am J Otolaryngol 1993;14:394-8.
(4.) Draf W, Weber R, Keen R, Constantinidis J. [Current aspects of frontal sinus surgery. I: Endonasal frontal sinus drainage in inflammatory diseases of the paranasal sinuses]. HNO 1995;43:352-7.
(5.) Gross CW, Gross WE, Becker DG. Modified transnasal endoscopic Lothrop procedure: Frontal drillout. Op Tech Otolaryngol Head Neck Surg 1995;6:193-200.
(6.) Har-El G, Lucente FE. Eadoscopic intranasal frontal sinusotomy. Laryngoscope 1995;105:440-3.
(7.) Hosemann W, Weber R, Keerl R, Lund V. Minimally Invasive Endonasal Sinus Surgery. New York: Thieme Medical Publishers, 2000.
(8.) Jacobs JB. 100 years of frontal sinus surgery. Laryngoscope 1997; 107:1-36.
(9.) Hosemann W. Kuhnel T. Held P. et al. Endonasal frontal sinusotomy in surgical management of chronic sinusitis: A critical evaluation. Am J Rhinol 1997; 11:1-19.
(10.) May M, Schaitkin B. Frontal sinus surgery: Endonasal drainage instead of an external osteoplastic approach. Op Tech Otolaryngol Head Neck Surg 1995;6:184-92.
(11.) Metson R. Endoscopic treatment of frontal sinusitis. Laryngoscope 1992;102:712-6.
(12.) Schaefer SD, Close LG. Endoscopic management of frontal sinus disease. Laryngoscope 1990;100: 155-60.
(13.) Weber R, Draf W, Keerl R, et al. [Long-term results of endonasal frontal sinus surgery]. HNO 1996;44:503-9.
(14.) Wigand ME, Hosemann WG. Endoscopic sinus surgery for frontal sinusitis and its complications. Am J Rhinol 1991;5:85-9.
(15.) Kuhn FA, Bolger WE, Tisdahl RG. The agger nasi cell in frontal recess obstruction: An anatomic, radiologic and clinical correlation. Op Tech Otolaryngol Head Neck Surg 1991;2:226-31.
(16.) Stammberger H. Endoscopic diagnosis and surgery of the paranasal sinuses and anterior skull base--the Messerklinger technique and advanced applications from the Graz school. Tuttlingen, Germany: Storz GmbH.
(17.) Goodale RL, Montgomery WW. Anterior osteoplastic frontal sinus operation. Five years' experience. Ann Otol 196l;70:860-80.
(18.) Weber R, Draf W, Constantinidis J. Osteoplastic microscopic frontal sinus surgery. Am J Rhinol 1994;8:247-51.
(19.) Weber R, Draf W, Kahle G, Kind M. Obliteration of the frontal sinus--state of the art and reflections on new materials. Rhinology 1999;37:1-15.
(20.) Weber R, Hosemann W, Draf W, et al. [Endonasal frontal sinus surgery with permanent implantation of a place holder]. Laryngorhinootologie 1997;76:728-34.
(21.) Shikani AH. A new middle meatal antrostomy stent for functional endoscopic sinus surgery. Laryngoscope 1994;104:638-41.
(22.) Toffel PH. Secure endoscopic sinus surgery with middle meatal stenting. Op Tech Otolaryngol Head Neck Surg 1995;6:157-62.
(23.) Amble FR, Kern EB, Ned B 3d, et al. Nasofrontal duct reconstruction with silicone rubber sheeting for inflammatory frontal sinus disease: Analysis of 164 cases. Laryngoscope 1996;106:809-15.
(24.) Neel HB, Whicker JH, Lake CF. Thin rubber sheeting in frontal sinus surgery: Animal and clinical studies. Laryngoscope 1976;86:524-36.
(25.) Neel HB 3d, McDonald TJ, Facer GW. Modified Lynch procedure for chronic frontal sinus diseases: Rationale, technique, and long-term results. Laryngoscope 1987;97:1274-9.
(26.) Stammberger H. [Complications of inflammatory paranasal sinus diseases including iatrogenically-induced complications]. Eur Arch Otorhinolaryngol Suppl 1993;1:61-102.
(27.) Hoyt WH 3d. Endoscopic stenting of nasofrontal communication in frontal sinus disease. Ear Nose Throat J 1993;72:596-7.
(28.) Clark RA, Henson PM. The Molecular and Cellular Biology of Wound Repair. New York: Plenum Press, 1988.
(29.) Cohen IK, Diegelmann RF, Lindblad WJ. Wound Healing: Biochemical and Clinical Aspects. Philadelphia: W.B. Saunders, 1992.
(30.) Hosemann W, Wigand ME, Gode U, et al. Normal wound healing of the paranasal sinuses: Clinical and experimental investigations. Eur Arch Otorhinolaryngol 1991;248:390-4.
(31.) Weber R, Keerl R, Huppmann A, et al. Investigation of wound healing after paranasal sinus surgery with time lapse video--first results. Am J Rhinol 1996;10:235-8.
(32.) Weber R, Keerl R, Jaspersen D, et al. Computer-assisted documentation and analysis of wound healing of the nasal and oesophageal mucosa. J Laryngol Otol 1996;110:1017-21.