Revision Sinus Surgery

aggregation and are formulated primarily for hemostasis, it would appear prudent to avoid leaving these materials in the cavity, as the fabric/platelet plug may be a matrix for fibrosis. In contrast, the Sinu-Foam dressing consists of high-viscosity, injectable cellulose-based material that can be instilled as a paste and may be more amenable to clearance though mucociliary flow and/or irrigation. Both animal and human studies are necessary to investigate the effects of these materials on mucosal healing after ESS.

Additional Considerations and Future Directions

Following application of an absorbable biomaterial in the form of a fabric or wafer, it is important that the agent is deliberately hydrated with saline. This allows the material

37 to convert to a more gel-like consistency that can be dissolved by mucociliary clearance and/or by the application of saline via spray or irrigation (Fig. 37.3). In reality, these “bioabsorbable” agents are more likely to dissolve rather than undergo true absorption by the surrounding tissues. Allowing the material to hydrate with ambient blood in the cavity may result in the formation of a plug consisting of fibrin and platelets, which may subsequently form a matrix for deposition of a collagen scar. In addition, allowing the material to persist in the cavity may result in frank incorporation of particles into the scar tissue or the regenerated mucosa, as studies have demonstrated [16]. This observation underscores the importance of ongoing postoperative management with endoscopic surveillance, debridement, and medical management.

Use of biomaterials does not apparently compensate for mucosal stripping during surgery, and in fact, murine studies have even demonstrated that hyaluronic acid ester is associated with significant osteoneogenesis if applied directly to denuded bone [11]. The use of absorbable biomaterials may be associated with significant patient comfort compared to a traditional pack and in some cases may augment healing, but does not replace the roles of mucosal preservation during surgery and meticulous postoperative care. It should also be stressed that no studies have demonstrated improved long-term outcome between the use of an absorbable and a traditional tampon pack; therefore, if there is potential for severe postoperative bleeding or if the middle turbinate is highly destabilized, use of a nonabsorbable spacer should be considered.

Future directions include conjugation or impregnation of the current molecules (i.e., hyaluronic acid) with antibiotics, anti-inflammatory medications (e.g,. steroids), compounds that dissolve biofilms, mediators that pro-

mote epithelialization (as was attempted with IGF-1), and those that prevent fibrosis (e.g., mitomycin-C). Furthermore, it is clear that agents composed of the same fundamental molecule, such as hyaluronic acid, may exhibit differential properties based on the material composition (polymerized as an ester versus carboxymethyl cellulose). The search goes on for the ideal preparation of the ideal molecule. In summary, these materials play a significant role in the armamentarium of the revision endoscopic sinus surgeon, but their strengths and shortcomings must be considered so the appropriate material can be selected for the given clinical scenario.

References

1.Asai K, Haruna S, Otori N, Yanagi K, Fukami M, Moriyama H (2000) Saccharin test of maxillary sinus mucociliary function after endoscopic sinus surgery. Laryngoscope 110:117–122

2.Bassiouny A, Abd El Raouf M, Atef A, Nasr S, Talaat S, Nasr M, Ayad E (2005) A comparative study between ciliary count and the degree of opacity of paranasal sinus CT scans in chronic rhinosinusitis pre and post FESS. J Laryngol Otol 119:950–954

3.Baumann A, Caversaccio M (2003) Hemostasis in endoscopic sinus surgery using a specific gelatin-thrombin based agent (FloSeal) Rhinology 41:244–249

4.Beck DE, Cohen Z, Fleshman JW, Kaufman HS, van Goor H, Wolff BG (2003) A prospective, randomized, multicenter, controlled study of the safety of Seprafilm adhesion barrier in abdominopelvic surgery of the intestine. Dis Colon Rectum 46:1310–1319

5.Chandra RK, Conley DB, Kern RC (2003) The effect of FloSeal on mucosal healing after endoscopic sinus surgery: a comparison with thrombin-soaked gelatin foam. Am J Rhinol 17:51–55

6.Chandra RK, Conley DB, Haines GK 3rd, Kern RC (2005) Long-term effects of FloSeal packing after endoscopic sinus surgery. Am J Rhinol 19:240–243

7.Cohen NA (2006) Sinonasal mucociliary clearance in health and disease. Ann Otol Rhinol Laryngol Suppl 196:20–26

8.Frenkiel S, Desrosiers MY, Nachtigal D (2002) Use of hylan B gel as a wound dressing after endoscopic sinus surgery. J Otolaryngol 31:S41–S44

9.Gall RM, Witterick IJ, Shargill NS, Hawke M (2002) Control of bleeding in endoscopic sinus surgery: use of a novel gelatin-based hemostatic agent. J Otolaryngol 31:271–274

10.Gunaydin S, Mccusker K, Vijay V (2005) Clinical performance and biocompatibility of novel hyaluronan-based heparin-bonded extracorporeal circuits. J Extra Corpor Technol 37:290–295

11.Jacob A, Faddis BT, Chole RA (2002) MeroGel hyaluronic acid sinonasal implants: osteogenic implications. Laryngoscope 112:37–42

12.Jameson M, Gross CW, Kountakis SE (2006) FloSeal use in endoscopic sinus surgery: effect on postoperative bleeding and synechiae formation. Am J Otolaryngol 27:86–90

13.Kuo MJ, Zeitoun H, Macnamara M, Wagstaff K, Carlin WV, Turner N (1995) The use of topical 5% lignocaine ointment for the relief of pain associated with post-operative nasal packing. Clin Otolaryngol Allied Sci 20:357–359

14.Lee JT, Kennedy DW, Palmer JN, Feldman M, Chiu AG (2006) The incidence of concurrent osteitis in patients with chronic rhinosinusitis: a clinicopathological study. Am J Rhinol 20:278–282

15.Longaker MT, Chiu ES, Adzick NS, Stern M, Harrison MR, Stern R (1991) Studies in fetal wound healing. V. A prolonged presence of hyaluronic acid characterizes fetal wound fluid. Ann Surg 213:292–296

16.Maccabee MS, Trune DR, Hwang PH (2003) Effects of topically applied biomaterials on paranasal sinus mucosal healing. Am J Rhinol 17:203–207

17.McIntosh D, Cowin A, Adams D, Wormald PJ (2005) The effect of an expandable polyvinyl acetate (Merocel) pack on the healing of the nasal mucosa of sheep. Am J Rhinol 19:577–581

18.Miller RS, Steward DL, Tami TA, Sillars MJ, Seiden AM, Shete M, Paskowski C, Welge J (2003) The clinical effects of hyaluronic acid ester nasal dressing (Merogel) on intranasal wound healing after functional endoscopic sinus surgery. Otolaryngol Head Neck Surg 128:862–869

19.Palmer J (2006) Bacterial biofilms in chronic rhinosinusitis. Ann Otol Rhinol Laryngol Suppl 196:35–39

20.Proctor M, Proctor K, Shu XZ, McGill LD, Prestwich GD, Orlandi RR (2006) Composition of hyaluronan affects wound healing in the rabbit maxillary sinus. Am J Rhinol 20:206–211

21.Rajapaksa S, McIntosh D, Cowin A, Adams D, Wormald PJ (2005) The effect of insulin-like growth factor 1 incorporated into a hyaluronic acid-based nasal pack on nasal mucosal healing in a healthy sheep model and a sheep model of chronic rhinosinusitis. Am J Rhinol 19:251–256

22.Rajapaksa SP, Cowin A, Adams D, Wormald PJ (2005) The effect of a hyaluronic acid–based nasal pack on mucosal healing in a sheep model of rhinosinusitis. Am J Rhinol 19:572–576

23.Shaw CL, Dymock RB, Cowin A, Wormald PJ (2000) Effect of packing on nasal mucosa of sheep. J Laryngol Otol 114:506–509

24.Shinkwin CA, Beasley N, Simo R, Rushton L, Jones NS (1996) Evaluation of Surgicel Nu-knit, Merocel and Vasolene gauze nasal packs: a randomized trial. Rhinology 34:41–43

25.Tarhan OR, Eroglu A, Cetin R, Y Nce A, Bulbul M, Altuntas YR (2005) Effects of seprafilm on peritoneal fibrinolytic system. ANZ J Surg 75:690–692

26.Tom LW, Palasti S, Potsic WP, Handler SD, Wetmore RF (1997) The effects of gelatin film stents in the middle meatus. Am J Rhinol 11:229–232

27.Woodworth BA, Chandra RK, Hoy M, Schlosser RJ, Gillespie MB (2006) SepraPack dressing after sinus surgery: a randomized trial. Arch Otolaryngol Head Neck Surg 135: P74–P75

28.Wormald PJ, Boustred RN, Le T, Hawke L, Sacks R (2006) A prospective single-blind randomized controlled study of use of hyaluronic acid nasal packs in patients after endoscopic sinus surgery. Am J Rhinol 20:7–10

4.The ethmoid vessels – assessing the relationship to the skull base and evidence of previous clip ligation.

5.Presence of Onodi cells (posterior ethmoid cells with pneumatization superior to the sphenoid) and a possible exposed optic nerve and susceptible skull base.

6.The internal carotid artery and the cavernous sinus with respect to the sphenoid sinus.

7.The degree of frontal sinus pneumatization with emphasis on anteroposterior dimension as well as distance from lamina to lamina. The presence of frontoethmoidal cells, the specific anatomy of the frontal recess, and the presence of septation.

Image guidance has been discussed in Chap. 29. It plays an increasingly important role in revision surgery and may reduce the complication rate [5, 11]. It is, however, no substitute for a thorough preoperative assessment, detailed knowledge of the anatomy, and a cautious respectful approach.

Pathological Factors

Current indications for open surgical approaches as described above largely incorporate recalcitrant inflammatory disease of the frontal sinus and less frequently the maxillary sinus, and then a variety of neoplasms including inverted papilloma, osteomas, and recurrent JNA. Prior to any revision surgery, maximal medical management needs to be instituted [3]. In order to do this effec-

tively, a clear understanding of the pathology is required. An individualized medical treatment plan should be formulated for each patient. Atypical facial pain should be further assessed and treated prior to any surgical intervention. Other disorders that cause facial pain should be considered [44]. A trial management of the neuropathic pain with carbamazepine or amitriptyline, should preceed surgical intervention [52]. This can also include patients who have small mucoceles after frontal sinus obliteration in order to distinguish symptomatic recurrences from atypical pain syndromes.

Identification and treatment of infection preoperatively aids revision surgery, reduces intraoperative blood loss and enhances postoperative care. Staphylococcus aureus and Gram-negative organisms such as Pseudomonas are recovered most commonly in patients presenting for revision surgery [32]. Both have been implicated in the pathogenesis of chronic rhinosinusitis either as a superantigen stimulus or in biofilm formation [36, 43].

Symptomatic mucoceles occurring after obliteration procedures require surgical intervention. These can be geographically removed from the natural sinus; image guidance may be required. Benign tumors recurring after failed open procedures may act differently from the primary tumor. The recurrent JNA is frequently fibrosed and may bleed less. Planes of dissection are more indistinct and invasion into vital structures may occur. The possibility of malignant transformation in inverted papillomas needs to always be carefully considered. Recurrence can occur directly onto the periorbita or dura.

340

Patients are counseled on the importance of continuing medical therapy before, during, and after their operation.

Surgical Techniques

Revising the C-L Procedure

Caldwell and Luc described the C-L operation more than 100 years ago as the surgical treatment for maxillary sinus disease [24]. During the last decades, less radical interventions using the endoscopic approach have mainly replaced the classical procedures performed for chronic and recurrent maxillary rhinossinusitis [2, 37]. Despite the success of endoscopic middle-meatus antrostomy coupled with the reports of fairly high morbidity rates with the C-L operation in the literature [4, 8, 16, 23, 30, 33, 40, 54], several indications remain, and still many surgeons advocate the use of this procedure for endoscopic

38 failures with irreversible mucosal disease [7, 25]. Comparative studies of endoscopic sinus surgery

and C-L have largely favored the endoscopic approach, demonstrating better ostial patency rates and overall patient symptomatic control [37, 49]. This has been attributed by some to the increased fibrosis and abnormal bony changes of the maxillary sinus that are encountered in a higher proportion of open procedures compared to

Raymond Sacks and Larry Kalish

endoscopic approaches (Fig. 38.2) [49]. However, select patients seem to benefit from radical removal of diseased and condemned mucosa through a C-L or one of its modified procedures, such as the canine fossa puncture [9, 10, 42].

The concepts that one needs to consider specific to the endoscopic revision of previously performed C-L surgery include:

1.It is not a physiological procedure.

2.The inferior meatal antrostomy does not surgically address the physiologic drainage tract of the maxillary sinus, which is directed to the middle meatus through the infundibulum and hiatus semilunaris [46].

3.The term “mucus recirculation phenomenon” is used to describe an antrostomy that did not accurately include the natural ostium of the maxillary sinus at the middle meatus and can predispose to chronic infection and possibly chronic osteitis (Fig. 38.3).

4.The natural ostium may be patent behind a completely intact uncinate; however, significant synechiae may lie just beyond, necessitating a wide appropriate antrostomy.

5.Synechiae and osteoneogenesis may result in abnormal barriers preventing access to all of the diseased mucosa in the maxillary sinus, and may interfere with disease monitoring.

6.With angled scopes and instrumentation, endoscopic repair of oroantral fistulae and retrieval of foreign

Fig. 38.3  Post-Caldwell-Luc mucus recirculation (arrow) from maxillary antrostomy to inferior antrostomy (not seen)

bodies is often safely and expediently achieved [22]. Following previous C-L procedures, the inferior antrostomy can be utilized for better access for scopes or instruments; however, synechiae and neo-osteogenesis can again slow progress and present potential traps.

7.The ethmoid complex is seldom addressed and recurrence from missed adjacent disease may need to be treated prior to revising the maxillary disease.

8.Postoperative debridement and monitoring is not always possible through the inferior antrostomy.

advertently into the maxilla. A ball-tipped, right-angled probe is passed through the inferior incision and slid anteriorly as close as possible to the uncinate’s insertion into the lateral nasal wall. We then palpate using the ball probe for synechiae resulting from the original C-L procedure. The probe is then pulled anteriorly, gently breaking down any synechiae and fracturing the uncinate at its insertion. A 45 °, upturned, through-cutting Blakesley forceps is used to cut the middle portion of the uncinate flush with the frontal process of the maxilla. The superior edge of the uncinate is then divided using a straight, through-cutting Blakesley forceps. A complete uncinectomy assists in exposing the natural maxillary ostium, accessory ostia, and surgically created ostia following the previous C-L procedure. A 45 °, angled endoscope is used while performing a wide maxillary antrostomy to confirm that the natural ostium has been incorporated into the surgical antrostomy, and to adequately assess the post-C-L antrum.

Careful assessment of preoperative imaging will direct the surgeon to potential areas of pathology behind newly formed bone and scar bands resulting from the previous C-L. Angled endoscopes, 45 ° and sometimes 75 °, together with angled microdebriders assist in locating and managing problem areas. Positioning instruments or endoscopes through a patent inferior antrostomy may also be useful. Care is taken to identify the infraorbital nerve and appreciate floor or orbit dehiscence. Pressure over the patient’s cheek will identify lateral sinus defects and aid in medializing pathological mucosa for debridement. The ethmoid sinuses are then addressed by performing complete anterior and posterior ethmoidectomies.

Endoscopic Revision of the C-L Procedure

The patient is appropriately positioned; the nose is then prepared with injection of lidocaine 2% + 1:100,000 epinephrine using a dental syringe and needle. The points of infiltration are at the axillae of the middle turbinate, the anterior end of the middle turbinate, and the lateral nasal wall. Neuropatties (1 × 1.5-inch) soaked with a mixture of 4% xylocaine solution, epinephrine 1:1000, and oxymetazoline are carefully placed into the sphenoethmoidal recess, lateral to the middle turbinate in the middle meatal recess, medial to the middle turbinate, and along the inferior turbinate.

An uncinectomy is performed by carefully identifying the free edge of the uncinate with a ball-tipped, rightangled probe, always being vigilant of potential synechiae and lamina dehiscence from the previous open surgery. Once the free edge is established, a Microfrance backbiting instrument is used to make the inferior cut of the uncinate. Care is taken to not to drop bony fragments in-

Revision of the External Ethmoidectomy

Although rarely used these days for inflammatory sinus disease, the external approach to the ethmoids is still encountered in the management of subperiosteal orbital abscesses, ligation of the ethmoid arteries, and as part of a skull-base approach [29, 35, 41]. Of particular concern when revising these procedures is medial orbital bone removal, which may allow the orbital contents to collapse into the ethmoid cavity, resulting in ethmoid and frontal recess obstruction (Fig. 38.4) [34]. Injury to the nasolacrimal duct occurs at a slightly higher rate with revision surgery following previous radical ethmoidectomy [20].

Endoscopic Revision of the External Ethmoidectomy

A complete uncinectomy and maxillary antrostomy is performed. Care is taken when performing the uncinectomy to avoid dehiscence of the lamina by gently palpat-

ing with a ball-tipped, right-angled probe. Intact lamina is identified as soon as possible and correlation made with preoperative imaging. The scrub assistant performs frequent balloting of the ipsilateral orbit so that any dehiscence is recognized early.

The ethmoid bulla may be intact after an open ethmoidectomy, but the surgeon should be cautious as dehiscent orbital contents may lie just beyond. We minimize the use of microdebriders and advocate careful curettage by placing the angled curette medial to the ethmoid bulla between the bulla and middle turbinate, palpating for a natural ostium, rotating the curette laterally, and pulling it anteriorly. Similar technique is used throughout the revision case to prevent inadvertent trauma to the periorbita or skull base.

The skull base is identified ideally after performing a sphenoidotomy and then followed anteriorly, this allows the surgeon to identify the skull base posterior to the previous surgery; this is important as the external ethmoidectomy can remove bone flush with the skull base, leading to the inadvertent breech when revising the case endoscopically.

Revision of the Frontal Sinus Trephine

Like many external procedures, the frontal trephine has evolved from a potentially cosmetically deforming procedure to a minimally invasive adjunct to endoscopic surgery [12, 28]. Its original description involved removal of enough bone to allow adequate visualization into the sinus and for placement of a drainage tube in order to

avoid mucosal necrosis, osteomyelitis, and intracranial complications [19]. When the frontal trephine fails, endoscopic frontal sinus surgery is performed as described in Chap. 14.

Revision of the Lynch-Howarth Procedure

In 1921, Lynch in the USA and Howarth in the United Kingdom popularized a procedure to remove the frontal sinus floor but with preservation of the anterior wall, which included removal of the frontal sinus mucosa, dealing with any intranasal obstruction and antral disease and creating a large frontonasal drainage via an ethmoidectomy [19]. Although multiple techniques to address the frontal recess have been described over the years, the Lynch-Howarth procedure has remained a viable alternative. In a severely scarred frontal recess with osteitic bone after multiple endoscopic surgical approaches, the Lynch-Howarth approach may still be a procedure worth considering, along with modified endoscopic Lothrop approaches and osteoplastic flap with or without obliteration (Fig. 38.5).

Endoscopic Revision

of the Lynch-Howarth Procedure

Positioning and preparation of the patient as described above. Image guidance is recommended if available. Careful assessment of the preoperative imaging is made to determine the best approach. The options usually include a

Draf IIb procedure or an endoscopic modified Lothrop procedure (Draf III; see below). This is often determined by the degree of orbital content prolapse, the presence of a middle turbinate, and the underlying pathology. For example, a frontal mucocele secondary to frontal recess obstruction with an intact middle turbinate may be adequately treated by a Draf IIb, whereas a recurrent frontal sinus osteoma in a patient with only a remnant middle turbinate will do better with a more aggressive endoscopic modified Lothrop procedure.

Osteoplastic Frontal Sinus Surgery with and Without Obliteration

This can be a useful procedure and can be associated with shortand long-term complications [1, 14, 21, 50]. Mucoceles can occur centrally within the obliterated sinus cavity and are separated from the nasal cavity by bone and thick fibrous tissue, resulting in no communication between the mucocele and the nasal cavity, and consequently no pathway for the surgeon to follow [51, 52]. Obliterated frontal sinuses are frequently smaller than their unobliterated counterparts, and adequate drainage is only achieved by drilling a wide communication, thereby creating a circumferential injury [47]. Surgeons have to allow for significant (30–55%) contracture of the neo-ostium, and physical space is limited especially in the anteroposterior dimension [21]

Fig. 38.5  a–d Coronal CT from anterior to posterior showing postoperative external Lynch-Howarth procedure to drain an anterolateral mucocele (indicated by “*” in d) caused by a frontal osteoma

Endoscopic Revision of the Failed Osteoplastic Flap Procedure with or Without Obliteration

Infiltration with lidocaine and epinephrine is performed above the axilla of the middle turbinate and on both sides of the midseptum adjacent to the middle turbinate. The soft tissue is removed with a microdebrider from the axilla of the middle turbinate to the roof of the nose, exposing the underlying bone. The microdebrider is utilized to remove all mucosa of the septum anterior to the attachment of the middle turbinate, and an area approximately 2.5 × 2.0 cm of mucosa is removed from both sides of the septum. The visible septal bone and cartilage is removed completely to create a wide window through which one can operate on both sides of the nose simultaneously. The septal window should be low enough that one can clearly see the axilla of the middle turbinate with a 0 ° telescope from the opposite nasal cavity. Frontal trephines have limited usefulness when an obliteration procedure has been performed, but in an osteoplastic procedure without obliteration there may be an indication to perform frontal trephine at this stage for fluorescein guidance.

The axilla of the middle turbinate on both sides is extended superiorly using a Hajek-Koeffler punch or a Kerrison rongeur. A 3.2-mm cutting burr is then used to remove the hardened frontal process of the maxilla above the axilla until a small amount of skin is exposed, thus defining the lateral anterior extent of the dissection. The procedure is performed on the opposite side and the

344

frontal sinus cavity is entered. The floor between the frontal sinus openings on both sides is removed with further removal of the beak of the frontal process of the maxilla. This limits the anterior dissection. It is important to note that the dissection is only performed in the superior and lateral direction until the frontal sinus cavity has been entered, so as not to damage the olfactory fossa and cause a cerebrospinal fluid (CSF) leak.

Once the frontal sinus has been entered on both sides, the dissection can now be brought medially until the intersinus septum is seen. The intersinus septum is then removed and at this stage, the frontal sinus opening forms a horseshoe appearance. Once this has been achieved, the frontal bone is removed until there is no anterior lip of bone separating the frontal sinus from the nasal cavity. The posterior projection is then thinned down until the first olfactory neuron is identified. Fat or any fibrous tissue needs to be gently removed from the frontal sinus cavity in order to create a continuous cavity from the frontal sinus into the nose. Malleable frontal curettes and

38 suction are critical in order to clear fat or other obliterative tissue from the sinus cavity. A 45 ° or 70 ° endoscope is typically required in order to achieve complete clearance of diseased tissue.

The use of packing material such as bismuth-iodo- form-paraffin paste or absorbable packs such as oxidized cellulose, is controversial and studies are currently underway to look at the long-term results of these packing materials.

Postoperative Care

Absorbable packing such as Surgicel Fibrillar (Ethicon) oxidized regenerated cellulose or Nasopore (Polyganics) biodegradable fragmentable foam, are preferred for postoperative hemostasis in order to maximize patient comfort and minimize adhesions and bleeding. Vigorous hypertonic, buffered-saline sinus irrigation using a bulb syringe or sinus rinse bottle, in the nasal douche position, is encouraged to optimize irrigation of the frontal recess and maxillary sinuses [53]. After frontal sinus surgery, the “mini trephines” are left in situ and 4-hourly flushes with dexamethasone, normal saline, and an antibiotic solution is given for 1–5 days postoperatively. Instruction sheets are given to the patient detailing postoperative care. Medical management is reinstituted on an individual basis and may include short course oral antibiotics, longer-term macrolide antibiotics, topical and systemic steroids, or recommencement of aspirin desensitization. All patients are seen in clinic 1 week postoperatively for endoscopic debridement of residual packing, crust, and division of any potential early adhesions, and then on a weekly or fortnightly basis for endoscopic surveillance

Raymond Sacks and Larry Kalish

and debridement until the surgeon is satisfied with the mucosal healing. Long-term follow up is also required and both surgeon and patient should remain vigilant for the signs of recurrence, as a good operative result should facilitate easier access to the sinuses for medical therapy.

Complications and Outcomes

Specific concerns during endoscopic revision following open surgery have been discussed herein. The surgeon should be aware of the increased risk of complication and remain vigilant at all times. Complications include:

1.CSF leak.

2.Orbital and optic nerve injury.

3.Major bleeding.

4.Nasolacrimal injury.

5.Anosmia.

6.Neural injury.

There have been very few papers directly addressing the outcome of endoscopic sinus surgery following open surgery. In a retrospective review, post-C-L failures had similar clinical improvements with either revision endoscopic surgery or revision C-L, 67% and 60%, respectively, and required similar number of revisions, 1.7±1.0 and 1.3±0.5, respectively [13]. However, the operative morbidity and potential sequelae of open surgery significantly favored an endoscopic approach.

Experimental studies have indicated that mucosal stripping leads to regeneration of mucosa that bears patchy, dysmorphic, and dyskinetic cilia [13, 14]. Thus, for endoscopic revision to be successful it depends on the regeneration of mucociliary function or residual mucosa left behind after the original open surgery. The persistence of maxillary sinus disease despite endoscopic evidence of complete uncinectomy and wide patency of the ostial outflow tract is likely to occur after complete mucosal stripping and may require a more aggressive endoscopic approach including an mega-antrostomy or an endoscopic medial maxillectomy approach [13].

Few studies have addressed endoscopic revision of open frontal sinus surgery including obliteration [15, 27, 48, 51]. A significant number of patients with persistent frontal sinusitis after obliteration can be successfully managed endoscopically. It is important to select patients carefully. Disease localized in the frontal recess or inferomedial frontal sinus is more likely to be successful than superior and lateral frontal disease, which may be best approached externally [15]. MRI and CT can be helpful in guiding the selection of the surgical approach. However, when radiological findings are equivocal, the surgeon must be aware of possible occult disease beyond the reach of endoscopic techniques. Patients undergoing en-