Revision Sinus Surgery

Fig. 18.5  a Coronal CT with extensive frontal and superior orbital wall erosion. Widened frontal recess and ostia. b Axial CT with extensive erosion of the posterior plate of the frontal sinus.

c Postoperative axial CT after endoscopic frontal surgery and removal of fungal debris. d Postoperative endoscopic view of the widened frontal ostium

Conclusions

It is important to consider that any one particular surgery may not suffice in the treatment of this aggressive disease. Medical therapy should be maximized prior to pursuing any revision surgery and should be directed at modulating the immune response to fungus. Surgery serves to eliminate the fungal debris present in the sinus cavities as well as to provide access for postoperative debridement and irrigations. Further harm during revision surgery should be conscientiously avoided. Revision surgery for AFRS is common; however, by following standard anatomical landmarks and by staging the procedure when needed, it can be performed safely with few complications.

References

1.Govindaraj S, Antunes M, Kennedy DW (2007) Revision sinus surgery. In: Kountakis SE, Onerci M (eds) Rhinologic and Sleep Apnea Surgical Techniques. Springer, Germany, pp 199–210

2.Mabry RL, Marple BF, Folker RJ, Mabry CS (1998) Immunotherapy for allergic fungal sinusitis: three years experience. Otolaryngol Head Neck Surg 119:648–651

3.Manning SC, Merkel M, Kriesel K, et al. (1997) Computed tomography and magnetic resonance diagnosis of allergic fungal sinusitis. Laryngoscope 107:170–176

 Core Messages

■While benign tumors can be safely and completely removed from the paranasal sinuses using endoscopic techniques, there is a generally infrequent but not negligible rate of recurrence.

■Long-term follow up with endoscopic and appropriate imaging surveillance is critical in recognizing early tumor recurrence.

■A high index of suspicion should be employed when there is a change in the endoscopic examination.

■Multimodality imaging is essential when there may be multiple sites of tumor recurrence.

■Intraoperative frozen section analysis is a vital step in assuring complete tumor removal at the time of revision surgery.

Introduction

Benign tumors of the nasal cavity and paranasal sinuses have been successfully removed utilizing transnasal endoscopic approaches for over the last two decades. With increased experience and skill, improved instrument design, and computer-aided surgery, an endoscopic approach has become the default choice for most rhinologic surgeons for the surgical management of sinonasal benign neoplastic diseases. Whether utilizing traditional open approaches or transnasal endoscopic techniques, there is an infrequent but not negligible rate of recurrent and residual tumor, which creates a unique clinical situation that deserves special discussion.

When performing revision surgery, the following questions should be answered:

1.Why did the previous surgery fail?

2.Was the nature/extent of disease underappreciated (Figs. 19.1 and 19.2)?

3.Were certain anatomic or imaging signs not recognized preoperatively?

4.Was intraoperative frozen section analysis utilized?

5.Were appropriate surgical techniques employed?

The answers to these questions will shed light on the reasons for failure and help assure success during subsequent surgery.

Fig. 19.1  Endoscopic view of recurrent pleomorphic adenoma of the nasal septum. The lesion was initially removed during a septoplasty at an outside facility. IT Inferior turbinate

While most benign tumors encountered in the nasal cavity and paranasal sinuses do not have malignant potential, the most commonly encountered one, inverted papilloma, is reported to undergo malignant degeneration in 5–15% of patients, with a working risk of 10% [6, 12]. For this reason, revision surgery for inverted papil-

Inverted papilloma is specifically the most commonly encountered Schneiderian papilloma and in general the most commonly encountered benign neoplasm of the nasal cavity and paranasal sinuses.

■ The recurrence of inverted papilloma ranges from 0 to 78% and has been related to surgical approach, extent of disease, and prior surgery.

When comparing open vs. endoscopic resection, recurrence rates for endoscopic approaches are favorable, and in some recent series, lower. Most of these studies, including one of the most recent, are retrospective reviews and meta-analyses [7]. However, as with most disease processes, the extent of disease is the single most important factor impacting recurrence. Lee et al. found the recurrence rate in their series of reoperations for inverted pap-

illoma to be 27.3% with Krouse stage III tumors, while lower stages did not recur [5]. Sautter et al. also found that extent of disease was the most important variable and that the surgical approach did not impact recurrence [10].

Preoperative Workup

When patients undergo surgery for inverted papilloma, regular endoscopic surveillance should be performed (Figs. 19.3 and 19.4). At times, early recurrent or residual disease can be identified and treated by simple excision in the office setting. Care should be taken to properly document the suspected site and the specimen sent to pathology. When there is suspicion for more advanced disease, radiographic evaluation should be performed to assist in determining the stage of disease. Computed tomography (CT) and magnetic resonance imaging (MRI) are both helpful in identifying the extent of disease and allow a preoperative staging of the patient’s recurrent disease. CT scan will also serve to demonstrate bony erosion or invasion into the skull base or the orbit. In patients with recurrent disease, there may be multiple distinct sites of neoplasia and a pure “stage” may not be determined. Prior surgery may also lead to postobstructive changes with mucocele formation, which cannot always be distinguished by CT alone. MRI is helpful in further distinguishing the nature of CT opacification (i.e., inspissated mucous vs. soft tissue). In addition to preoperative planning, this also has implications for prognosis (Figs. 19.5 and 19.6).

Fig. 19.3  Endoscopic view of recurrent inverted papilloma on the right lamina papyracea

■Areas of hyperostosis/osteoneogenesis on CT should be identifiedasthishasbeenshowntohighlycorrelatewith site of origin of inverted papilloma (Fig. 19.7b) [4, 14].

Comparison with CT imaging taken prior to the initial surgical procedure, if available, will be helpful in this setting as new areas of hyperostosis may be the result of prior surgery and not necessarily recurrent tumor origin.

Fig. 19.4  Endoscopic inspection of the right lamina papyracea 2 years postresection of recurrent inverted papilloma

When CTs are obtained, a surgical navigation protocol should be employed so that this technology can be utilized during the revision surgical procedure.

Currently, there are several staging systems for inverted papilloma, with that proposed by Krouse used most commonly (Table 19.1) [8]. Citardi et al. have proposed a new classification that eliminates malignancy, which was included in the Krouse system (Table 19.2) [1]. Regardless

Fig. 19.5  Coronal CT showing opacified sphenoid sinus with extensive pterygoid recess pneumatization in a patient with posterior ethmoid sinus inverted papilloma

Fig. 19.6  T1-weighted coronal magnetic resonance image showing intermediate signal intensity in the sphenoid sinus

of the system chosen, the most effective approach is to assimilate the endoscopic and imaging data in order to counsel the patient as to the appropriate choice of procedure as well as prognosis on an individual basis.

Once the preoperative stage and the extent of disease have been determined, a plan for revision surgery should be made. Patients are instructed to discontinue medications that would lead to unnecessary bleeding at the time of surgery. Decreased visualization related to a bloody field can lead to increased risk for complications and incomplete surgery resulting in residual/recurrent disease. It is important to ask specifically about the use of nonpre-

scription medications (nonsteroidal anti-inflammatory drugs, aspirin, vitamin E) and herbal products (green tea, Ginkgo biloba), which are not often disclosed but can lead to troublesome bleeding. Discontinuation of prescription anticoagulants is best managed after discussion with the prescribing physician regarding the risks of medical complications associated with their cessation.

Commonly encountered risks of endoscopic sinus surgery should be discussed with each patient. Specific risks related to the extent of proposed surgery should be further disclosed. For example, removal of a recurrent osteoma along the ethmoid roof may require/involve removing a

portion of the ethmoid roof, resulting in an anticipated cerebrospinal fluid leak and its attendant repair.

■Depending on the location of the surgery (hospital vs. ambulatory surgery center), special arrangements may need to be made to ensure that intraoperative frozen section analysis is available as an integral part of the revised surgical procedure.

Surgical Technique

Once the patient is brought to surgery, maximal decongestion of the nasal mucosa is started in the preoperative holding area by applying topical decongestants. In the operating room, the choice of general anesthetic agents is important. Wormald et al. have proposed total intravenous anesthesia as a method by which the peripheral vasodilatation associated with certain inhalation agents is avoided. The importance of controlling the heart rate in addition to blood pressure is emphasized in this technique [13]. Once appropriately anesthetized, local infiltration of vasoconstrictors is carried out along the lateral nasal wall and in the region of the sphenopalatine artery, either transnasally or transpalatally via the greater palatine foramen. Finally, topical vasoconstrictors are applied to the nasal mucosa using cotton pledgets. While these

steps may be time consuming, they will lead to improved hemostasis, which makes the surgical procedure much less problematic.

As an initial step, secretions and crusts are removed from the operative site. Suspicious areas should be sampled and sent to the pathologist for frozen section analysis (Fig. 19.8). Care must be taken to document these sites as there may be multiple separate foci of tumor that needs to be addressed. En bloc resection is preferred but not always possible in revision surgery. All gross tumor must be removed along with surrounding, normal-appearing tissue to assure complete resection. The permanent specimen should be clearly marked for orientation. Intraoperative frozen section analysis of surgical margins is important and will help dictate further resection.

■In general with benign disease, natural barriers such as the lamina papyracea and dura should not be removed unless there is obvious tumor involvement.

Areas of hyperostosis should be carefully reduced with a diamond bur on the microdebrider. Computer-aided surgery is helpful in these instances in gauging progress (Fig. 19.9). Depending on the location and extent of disease, the operating surgeon should be comfortable with advanced endoscopic techniques, such as the endoscopic modified Lothrop procedure and the transmaxillary/

transpterygoid approach to an extensively pneumatized sphenoid sinus. Hemostasis is achieved in the surgeon’s preferred fashion and nasal packing is placed as indicated.

Other Benign Neoplasms

Fibro-osseous Lesions

19 Osteoma, ossifying fibroma, and fibrous dysplasia constitute the range of fibro-osseous lesions affecting the paranasal sinuses. It is generally agreed that symptomatic lesions should be removed and complete resection is likely with osteoma and ossifying fibroma. However, complete removal is rarely achievable with fibrous dysplasia as it usually develops along existing bony planes.

■Sacrifice of vital structures is not appropriate when treating benign neoplasia.

Tumor recurrence of osteoma and ossifying fibroma is related to incomplete resection. At times the operating surgeon may elect to leave a “shell” of tumor along the skull base or lamina papyracea during primary surgery to avoid the risk of intracranial or intraorbital injury associated with complete tumor removal. The indications

Fig. 19.9  Intraoperative view after tumor resection and reduction of the hyperostotic region with a diamond bur

for revision surgery are no different than for primary surgery. Removal of symptomatic recurrent lesions is appropriate and complete tumor removal should be the goal (Figs. 19.10 and 19.11). Careful use of powered burs is helpful in reducing the overall bulk of these lesions. As with primary surgery, identifying a cleavage plane is helpful in separating the base of the lesion from the surrounding site of attachment. The use of a curette or osteotome enables the surgeon to safely remove the tumor while avoiding injury to underlying structures. The potential for orbital and intracranial complications should be discussed in advance with the patient, including intraoperative plans for repair and their impact on recovery should they occur.

Angiofibroma

Angiofibroma is relatively uncommon and recurrence, as with other tumors, is often related to the extent of disease at the time of initial surgery.

■Angiofibroma skull-base invasion was associated with recurrence in 27.5% of patients [3].

Herman et al. urged that all patients, including those with asymptomatic tumor remnants, be followed closely with

Fig. 19.10  Intraoperative view of recurrent left frontal sinus osteoma

serial CT [3]. Some small remnants may involute, while those that become symptomatic can be treated with radiation therapy or revision surgery. Scholtz et al. found as low as 15% recurrence in 14 patients over an 11-year period [11]. Seven patients underwent a transnasal endoscopic approach, which they conclude to be appropriate in patients with lesions that involve the nasopharynx, nasal cavity, paranasal sinuses, and pterygopalatine fossa. Preoperative arteriography with embolization has been shown to reduce intraoperative blood loss [2, 9].

Conclusion

Benign tumors presenting in the nasal cavity and paranasal sinuses can be safely and successfully removed using transnasal endoscopic techniques. Success has also been achieved when tumors have extended beyond the confines of the nasal cavity and paranasal sinuses. The single most important factor in tumor recurrence is the extent of disease at initial presentation. Patients should be followed closely long term to enable early detection of recurrent disease. Questionable findings on endoscopic surveillance should be viewed with a high index of suspicion and appropriate biopsy and/or imaging studies obtained. Depending on the extent of tumor and the skills of the operating surgeon, endoscopic techniques once again can be safely used to address recurrent disease.

Fig. 19.11  Intraoperative view of osteoma specimen as it was resected

References

1.Cannady SB, Batra PS, Sautter NB, Roh HJ, Citardi MJ (2007) New staging system for sinonasal inverted papilloma in the endoscopic era. Laryngoscope 117:1283–1287

2.Economou TS, Abemayor E, Ward PH (1998) Juvenile nasopharyngeal angiofibroma: an update of the UCLA experience, 1960–1985. Laryngoscope 98:170–175

3.Herman P, Lot G, Chapot R, Salvan D, Huy PI (1999) Long term follow-up of nasopharyngeal angiofibromas: analysis of recurrences. Laryngoscope 109:140–147

4.Lee DK, Chung SK, Dhong HJ, Kim HY, Kim HJ, Bok KH (2007) Focal hyperostosis on CT of sinonasal inverted papilloma as a predictor of tumor origin. AJNR Am J Neuroradiol 28:618–621

5.Lee TJ, Huang SF, Lee LA, Huang CC (2004) Endoscopic surgery for recurrent inverted papilloma. Laryngoscope 114:106–112

6.Lesperance MM, Esclamdo RM (1995) Squamous cell carcinoma arising in inverted papilloma. Laryngoscope 105:178–183

7.Mirza S, Bradley PJ, Acharya A, Stacey M, Jones NS (2007) Sinonasal inverted papillomas: recurrence, and synchronous and metachronous malignancy. J Laryngol Otol 121: 857–864

8.Krouse JH (2000) Development of a staging system for inverted papilloma. Laryngoscope 110:965–968

166

9.Onerci TM, Yücel OT, Oğretmenoğlu O (2003) Endoscopic surgery in treatment of juvenile nasopharyngeal angiofibroma. Int J Pediatr Otorhinolaryngol 67:1219–1225

10.Sautter NB, Cannady SB, Citardi MJ, Roh HJ, Batra PS (2007) Comparison of open versus endoscopic resection of inverted papilloma. Am J Rhinol 21:320–323

11.ScholtzA,AppenrothE,Kammen-ollyK,ScholtzLU,Thum- fart WF (2001) Juvenile nasopharyngeal angiofibroma: management and therapy. Laryngoscope 111:681–687

Michael J. Sillers and Yvonne Chan

12.von Buchwald C, Bradley PJ. Risks of malignancy in inverted papilloma of the nose and paranasal sinuses (2007) Curr Opin Otolaryngol Head Neck Surg 15(2):95–98

13.Wormald PJ, van Renen G, Perks J, Jones JA, LangtonHewer CD (2005) The effect of the total intravenous anesthesia compared with inhalational anesthesia on the surgical field during endoscopic sinus surgery. Am J Rhinol 19:514–520

14.Yousuf K, Wright ED (2007) Site of attachment of inverted papilloma predicted by CT findings of osteitis. Am J Rhinol 21:32–36

 Core Messages

■Endoscopic repair of cerebrospinal fluid (CSF) leaks and meningoencephaloceles is largely successful, with rates of skull-base defect closure greater than 90%.

■Certain factors may increase the potential for recurrence of CSF leak or meningoencephalocele. Identification of such factors in the preoperative period allows the surgeon to alter treatment protocols accordingly.

■Diagnosis of recurrent CSF leaks and meningoencephaloceles often incorporates preoperative and intraoperative techniques. A combination of laboratory testing, radiologic imaging, and special procedures may be required to accurately diagnose the site of a skull-base defect.

■Surgical techniques for endoscopic repair of recurrent CSF leaks and meningoencephaloceles vary by the site and size of the skull-base defect, as well as surgeon experience. Certain types of recurrent CSF leaks and meningoencephaloceles may require additional perioperative measures for increased repair success.

Introduction

In 1926, Dandy reported the first successful intracranial repair of a cranial defect in a patient with pneumocephalus [9]. Extracranial repair of a skull-base defect and cerebrospinal fluid (CSF) leak via naso-orbital incision was subsequently reported in 1948 by Dohlman [10]. The first report of endonasal endoscopic skull-base defect repair occurred in a case of iatrogenic skull-base injury and was published by Wigand in 1981 [53]. Following this, Mattox and Kennedy described the techniques for endoscopic repair of CSF leaks in detail and expanded the indications

Contraindications for Transnasal Endoscopic

Repair of CSF Leaks and Meningoencephaloceles    168 Preoperative Workup  . . . . . . . . . . . . . .   168

to skull-base defects caused by craniofacial trauma and meningoencephaloceles [24].

In present times, success rates for closure of skull-base defects by endoscopic techniques remain greater than 90% in most series [6, 8, 15, 17–19, 21–24, 26, 29, 56], as compared to traditional open skull-base defect repair via craniotomy, with success rates reported in the range of 60–80% [1, 16, 34]. In addition, endoscopic transnasal