Revision Sinus Surgery

Fig. 26.5  Coronal CT with Onodi cell (white arrow); note the proximity of the optic nerve to roof of the Onodi cell

erly calibrated, can provide additional reinforcement of the location of the skull base and the depth of dissection. Routinely entering the sphenoid sinus in its inferiorand medial-most aspect continues to be the safest means by which to enter the sphenoid sinus, providing the greatest distant between any instrumentation and the superior, lateral course of the optic nerve.

Unfortunately, damage to the optic nerve is unlikely to be forgiving. Severing or disrupting the fibers of the optic nerve will more than likely result in permanent visual compromise. Heat damage or swelling related to manipulation of an area around the optic nerve sheath can be treated with high-dose oral steroids and/or an optic nerve decompression (however, this is usually unsuccessful). An optic nerve decompression relies upon endoscopic release of the optic sheath along its entire course to the annulus of Zinn [15]. In either case, consultation with ophthalmology and documentation of the patient’s vision immediately post-injury and after any intervention is essential. Postoperative monitoring in a unit where routine monitoring of vision can be accomplished is essential.

Carotid Injury

■To avoid injury to the carotid artery in revision sinus surgery:

1.Recognize that carotid dehiscence occurs up to 25% of the time.

2.Avoid powered instrumentation or biting/grasping instrumentation within the sphenoid sinus (espe-

cially posterior and lateral) without proper visualization.

The potentially catastrophic complication of ICA injury can result in life-threatening hemorrhage, hemispheric stroke, coma, or death [33]. The anatomic relationship of the cavernous portion of the ICA in the sphenoid sinus results in a potential risk of injury during endoscopic sinus surgery. Ascending toward the posterior clinoid process and passing forward by the side of the body of the sphenoid bone, the ICA curves upward and perforates the dura mater, forming the roof of the sphenoid sinus. It is in this lateral and posterior aspect of the sphenoid sinus that iatrogenic injury can occur.

As one enters the sphenoid sinus, the ICA can be dehiscent approximately 20–25% of the time along some portion of its course [34]. Entering in the lateral, posterior aspect of the sphenoid sinus with a blunt instrument can pierce the uncovered ICA. In addition, blindly placing a powered debrider or other grasping/biting instrument into the posterior and lateral aspect of the sphenoid sinus may result in carotid artery damage.

Visualization, preparation, and finesse will aid the surgeon in preventing this potentially lethal complication. Identifying a preexisting dehiscence when examining the axial computed tomography images will alert the surgeon to a potential hazard. Proper visualization and location of the natural os of the sphenoid sinus as it has been described will also aid in prevention. As mentioned earlier, the safest method by which to widen the sphenoid sinus is identification of the natural os and opening the sinus in its medialand inferior-most aspect. Avoiding placement of any powered instrumentation or grasping/biting instrument in the posterior and lateral aspect of the sphenoid sinus is highly recommended. Routine placement of an oropharyngeal throat pack in any patient having sphenoid or sinus surgery offers the benefit of potential complete oronasopharyngeal tamponade in the case of carotid injury and uncontrolled bleeding.

Identification of a compromise to the integrity of the ICA is generally not difficult. Occasionally, however, subendothelial damage to the ICA may result in a delayed aneurysm, pseudoaneurysm, or delayed rupture of the carotid artery [26]. For small, pinhole-type injuries where visualization can be accomplished, placement of small piece of Gelfoam or an epinephrine-soaked pledget over the area can allow for clot formation. Gentle packing around this area and use of fibrin sealants can also help control a small manageable defect. Notably, vasospasm of the ICA can still occur with this manipulation, resulting in hemispheric stroke [22].

More significant bleeding, however, will quickly obscure endoscopic visualization. In such a case, immediate packing of the nasal cavity will be required. Placement of

230

a Foley catheter or other into the posterior nasal pharynx and inflating the balloon, in combination with providing counterpressure to the balloon anteriorly allows for tight packing within the nasopharynx, tamponading the hemorrhage. With an oropharyngeal throat pack in place, any significant nasopharyngeal packing can provide the required pressure to slow or halt a carotid bleed. Immediate transfer to an institution/location where endovascular procedures are performed is a necessity. A neurovascular interventional radiologist may be able to control such damage with endovascular coiling or other hemostatic measures. Such procedures can result in hemispheric stroke or blindness; however, this risk is overshadowed by the alternative – a fatal hemorrhage [22].

After control of the hemorrhage is achieved, neurological assessment must be accomplished and documented. Involvement of the neurologist is a necessity. Discussions with the patient’s family as to the seriousness of this potentially lethal complication should be undertaken. Observation in an appropriate hospital unit is a necessity.

CSF Leak/Skull-Base Penetration

■To prevent and manage anterior skull-base injuries:

1.Recognize high-risk areas of the fovea ethmoidalis, cribriform plate, and the anterior, medial, superior sphenoid sinus.

2.Avoid operating medially, superior to the middle turbinate.

3.Recognize the presence of CSF as a warning sign that the skull base has been penetrated.

Clearly, the anatomy of the paranasal sinuses places the endoscopic sinus surgeon millimeters from the intracranial contents of the anterior cranial fossa. The roof of the nasal cavity represents the floor of the anterior skull base. The thin lateral lamella connecting the cribriform plate with the fovea ethmoidalis provides a fragile barrier between the nasal cavity and the intracranial contents. Posteriorly, the sloping extension of the planum sphenoidale represents a potential entry point into the intracranial cavity.

Injury to the skull base in any of theses areas can occur in a variety of ways. Initially, failure to identify the level of the skull base at any point can enable puncture of the skull base with sharp, dull, or powered instrumentation [13]. An ethmoidectomy carried too far medial and superior can result in penetration of the lateral lamella of the cribriform plate or fovea ethmoidalis. It should be mentioned that the medial fovea ethmoidalis is up to

26 ten times thinner than the lamina papyracea. Aggressive management of a middle turbinate that inserts into the lateral lamella can result in a fracture of the skull base in this area and resultant CSF leak. Entrance into the sphe-

John Scianna and James Stankiewicz

noid sinus or Onodi-type cell medially in too high a manner can also result in skull-base penetration. With regard to revision surgery, the high-speed cutting drills used in revision frontal sinus surgery can easily be misplaced or skip into areas adjacent to the frontal sinus recess, resulting in skull-base injury [28].

Prevention begins with careful review and identification of the skull base on preoperative imaging. Evaluation for preexisting dehiscence and determination of the level of the skull base is mandatory. Three-dimensional imaging can aid in this endeavor. However, basic coronal imaging is generally used for determining the skull-base level. Multiple classifications have been used to describe the height of the skull base. While the measurements of the Keros classification can be extremely useful, a practical means of evaluating the skull base can be accomplished by comparing the level of the cribriform to the orbit [3]. A high skull base is at the superior aspect of the orbit, whereas a low skull base is located at one-third to one-half of the horizontal plane into the orbit. Any skull base at the level of the medial rectus or below is hazardous (Fig. 26.6). In addition to understanding the height associated with the skull base, understanding that in some individuals the fovea ethmoidalis slopes toward the cribriform plate, whereas in other individuals the fovea ethmoidalis is more vertical forming a “goblet” shape (Fig. 26.7) [10].

Beyond preoperative review of imaging, care must be taken during revision surgery not to accidentally puncture the skull base. Computer image guidance can be used to confirm the level of the skull base, understanding that visualization of the skull base as it slopes away from the surgeon supersedes computer imaging. In addition, the posterior level of the skull base is generally no more than 7 cm from the nasal sill and slopes in an arc anteriorly. Care must be taken in the areas of the superior aspect of the sphenoid sinus, the cribriform plate, and the fovea ethmoidalis. In revision cases, the middle turbinate may be absent. If present, one must avoid operating superiorly medial to the middle turbinate as this is a key “danger” zone and can easily lead to skull-base penetration. addition, scraping, using a bone rongeur, or aggressively cleaning the skull base in a variety of ways can result in a CSF leak. Understanding that sinus disease does not require resection to the skull-base mucosa, and allowing for a 2- to 3-mm cushion zone, even in revision cases, should adequately ensure avoidance of skull-base penetration, CSF leak, and complication.

CSF leak is the first sign that the skull base has been penetrated; it also represents a clear warning not to proceed any further with powered or other instrumentation. The endoscopic sinus surgeon should be well versed in the appearance of the skull base and recognize the white color of the dura. CSF rhinorrhea seen at the time of surgery is generally evident as gush of clear fluid that washes

232

away blood from the area. This is known as the “washout sign” [2]. Immediate CSF rhinorrhea may be short lived in that as CSF stores are depleted or as cerebral tissue prolapses into the defect, the rhinorrhea may resolve.

When it is evident that the skull base has been penetrated, intervention must occur. While with skullbase trauma, approximately 95% of the time CSF rhinorrhea is evident within the 3 months of injury, there is no literature to date that outlines the time line for identifying skull-base injury as a result of only endoscopic-sinus- surgery-induced trauma [25]. Regardless of the timing of injury, repair of the skull base is necessary. Skull-base dehiscence and CSF rhinorrhea unrecognized or untreated can result in chronic headache, pneumocephalus, meningitis, coma, and even death.

Multiple methods of skull-base repair exist. A variety of materials can be used to graft a skull-base injury: synthetic dura materials, septal mucosa, septal or turbinate cartilage, temporalis fascia/muscle, fat, and vascularized tissue such as rectus abdominus. Additionally, on-lay, over-lay, under-lay, and sandwich grafts have all been described as appropriate techniques for endoscopic repair of a skull-base defect. A traditional open, anterior cranial approach to the skull base can also be considered in cases where endoscopic repair fails. Endoscopic repair, however, is associated with a high success rate (approximately 90%) and considerably lower mortality [6]. Associated with repair of a skull-base defect, the use of CSF-reducing medications such as acetazolamide, and the use of lumbar drains have been advocated. In addition, with leaks found in a delayed fashion, the off-label use of intrathecal fluorescein at a concentration of one-tenth of 1 ml mixed with 10 ml of preservative free 0.9% saline or the patient’s own CSF can aid in the localization of a defect. While not

26

Fig. 26.8  Fluorescein-dyed cerebrospinal fluid (CSF) leak

John Scianna and James Stankiewicz

essential and requiring a separate consent indicating that a risk of seizure and/or chemical meningitis is associated with intrathecal fluorescein injection, its use can be quite helpful in isolating the area of a CSF leak (Fig. 26.8, Video 26.1) [25].

In addition to repair, a consultation with neurosurgeon is indicated, especially if a lumbar drain is contemplated. If intraoperative skull-base penetration is noted, discussion with the anesthetist about avoiding positive-pressure ventilation, which may result in pneumocephalus, should occur. The following measures should also be considered: no heavy lifting/strenuous activity, avoidance of straws, avoidance of nose blowing, and an initial period (24 h) of strict bed rest at minimal (< 15°) incline with gradual increase to a 90° position. If cortical injury is suspected, discussions with a neurosurgeon and neurologist as appropriate are required together with obtaining appropriate radiographic imaging. A period of observation in a monitored setting of at least 24 h should be provided.

Conclusion

All surgery is fraught with potential complications. Endoscopic sinus surgery as described herein involves highrisk territory with multiple vital structures in the immediate vicinity. Revision sinus surgery poses significant additional risks including operating in a field with potentially dehiscent structures, lack of reliable landmarks, and scarring, which can result in decreased visualization secondary to general oozing or bleeding. Understanding these risks, frank discussions with patients including well-documented informed consent, proper preoperative evaluation, radiographic evaluation, and intraoperative evaluation are mandatory. Avoidance of complications remains the best treatment. However, early recognition of a complication can help prevent a minor complication from becoming a major complication. The ability to appropriately intervene when a complication becomes evident relies on surgical skill, anatomic knowledge, and experience. Involving appropriate colleagues and specialists can be an invaluable asset. Observation and diligence postcomplication is essential. Finally, open discussions with patients and their families can help prevent legal sequelae associated with surgical complications.

Tips/pearls to avoid complications in revision sinus surgery:

1.Always obtain a detailed informed consent.

2.Always obtain appropriate preoperative imaging.

3.Computer image guidance can be helpful in revision surgery.

4.Always review all available imaging and previous operative reports and have them available in the operating suite.

5.Recognize areas of potential complication prior to surgery.

6.Maintain visualization throughout all endoscopic procedures.

7.Identify key landmarks when available: middle turbinate, maxillary antrostomy, sphenoid os, nasopharynx.

8.Recognize complications as early as possible.

9.Intervene appropriately and definitively when complications occur.

10.Involve appropriate colleagues and experts, transferring patients when necessary.

11.Always discuss complications, treatments, and required interventions with patients and their family.

References

1.Buus DR. Farris BK (1990) Ophthalmic complications of sinus surgery. Ophthalmology 97:612–619

2.Cumberworth VL, Sudderick RM, MacKay IS (1994) Major complications of functional endoscopic sinus surgery. Clin Otolaryngol 19:248–253

3.Gauba V, Saleh G, Dua G, et al. (2006) Radiological classification of anterior skull base anatomy prior to performing medial orbital wall decompression. Orbit 25:93–96

4.Graham SM, Nerad JA (2003) Orbital complications in endoscopic sinus surgery using powered instrumentation. Laryngoscope 113:874–878

5.Hackman TG, Ferguson BJ (2005) Powered instrumentation and tissue effects in the nose and paranasal sinuses. Curr Opin Otolaryngol Head Neck Surg 13:22–26

6.Hegazy HM, Carrau RL, Snyderman CH, et al. (2000) Transnasal endoscopic repair of cerebrospinal fluid rhinorrhea: a meta-analysis. Laryngoscope 110:1166–1172

7.Holtz L (2006) Wins, Losses and Lessons: an Autobiography. Harper Collins, New York

8.Huang CM, Meyer DR, Patrinely JR, et al. (2003) Medial rectus muscle injuries associated with functional endoscopic sinus surgery: characterization and management. Ophthal Plast Reconstr Surg 19:25–37

9.Jiang RS, Hsu Cy (2002) Revision functional endoscopic sinus surgery. Ann Otol Rhinol Laryngol 111:155–159

10.Jones TM, Almahdi JM, Bhalla RK, et al. (2002) The radiological anatomy of the anterior skull base. Clin Otolaryngol 27:101–105

11.Kim HJ, Kim CH, Song MS, Yoon JH (2004) Diplopia secondary to endoscopic sinus surgery. Acta Otolaryngol 124:1237–1239

12.Kim HU, Kim SS, Kang SS, et al. (2001) Surgical anatomy of the natural ostium of the sphenoid sinus. Laryngoscope 111:1599–1602

13.Lee JC, Song YJ, Chung YS, et al. (2007) Height and shape of the skull base as risk factors for skull base penetration during endoscopic sinus surgery. Ann Otol Rhinol Laryngol 116:199–205

14.Lund VJ, Wright A, Yiotakis J (1997) Complications and medicolegal aspects of endoscopic sinus surgery. J Roy Soc Med 90:422–428

15.Luxenberger W, Stammberger H, Jebeles JA, Walch C (1998) Endoscopic optic nerve decompression: the Graz experience. Laryngoscope 108:873–882

16.Maniglia AJ (1991) Fatal and other major complication of endoscopic sinus surgery. Laryngoscope 101:349–354

17.May M, Schiatkin B, Kay SL (1994) Revision endoscopic sinus surgery: six friendly surgical landmarks. Laryngoscope 104:766–767

18.McDonald SE, Robinson PJ, Nunez DA (2007) Radiological anatomy of the anterior ethmoidal artery for functional endoscopic sinus surgery. J Laryngol Otol 122:264–267

19.McInnes G, Howes DW (2002) Lateral canthotomy and cantholysis: a simple, vision-saving procedure. Can J Emerg Med 4:49–52

20.Metson R, Pletcher SD (2006) Endoscopic and optic nerve decompression. Otolaryngol Clin North Am 39:551–561

21.Meyers RM, Valvassori G (1998) Interpretation of anatomic variation of computed scan of the sinuses: a surgeon’s perspective. Laryngoscope 108:422–425

22.Park AH, Stankiewicz JA, Chow J, Azar-Kia B (1998) A protocol for management of a catastrophic complication of functional endoscopic sinus surgery: internal carotid artery injury. Am J Rhinol 12:153–158

23.Rene C, Rose G, Lenthall R, Moseley I (2001) Major orbital complications of endoscopic sinus surgery. Br J Ophthalmol 85:598–603

24.Sadoughi B, Brown SM, Nachlas NE, Fried MP (2006) Image guided surgery. E-medicine. Available at http://www. emedicine.com/ent/topic396.htm

25.Scianna JM, Stankiewicz JA (2006) CSF Rhinorrhea. E- medicine. Available at http://www.emedicine.com/ent/ topic332.htm

26.Sood S, Timothy J, Anthony R, et al. (2000) Extracranial internal carotid artery pseudoaneurysm. Am J Otolaryngol 21:259–262

27.Stankiewicz JA (1989) Blindness and intranasal ethmoidectomy: prevention and management. Otolaryngol Head Neck Surg 101:322–329

28.Stankiewicz JA (1991) Cerebrospinal fluid fistula and endoscopic sinus surgery. Laryngoscope 101:250–256

29.Stankiewicz JA, Chow JM (1999) Two faces of orbital hematoma in intranasal (endoscopic) sinus surgery. Otolaryngol Head Neck Surg 120:841–846

234

30.Stankiewicz JA (1987) Complications of endoscopic nasal surgery: occurrence and treatment. Am J Rhinol 1:45–49

31.Streitmann MJ, Otto RA, Sakai CS (1994) Anatomic considerations in complications of endoscopic and intranasal surgery. Ann Otol Rhinol Laryngol 103:105–109

32.Thanaviratananich S, Chaisiwamongkol K, Kraitrakul S, et al. (2003) The prevalence of an Onodi cell in adult Thai cadavers. Ear Nose Throat J 82:200–204

John Scianna and James Stankiewicz

33.Weidenbecher M, Huk WJ, Iro H (2005) Internal carotid artery injury during functional endoscopic sinus surgery and its management. Eur Arch Otorhinolaryngol 262:640–645

34.Yanagisawa E (2002) The optic nerve and internal carotid artery in the sphenoid sinus. Ear Nose Throat J 81:611–612

35.Yung CW, Moorthy RS, Lindley D, et al. (1994) Efficacy of lateral canthotomy and cantholysis in orbital hemorrhage. Ophthalmol Plast Reconstr Surg 10:137–141

 Core Messages

■A small and inappropriately placed bony opening, inadequate and improperly opened rhinostomy, excessive scar tissue production, anatomical abnormalities, and concomitant paranasal sinus infections are the most important causes for the failure of dacryocystorhinostomy (DCR) surgery.

■The endonasal approach is well suited for revision DCR surgery because the residual lacrimal sac can be directly accessed through the previous bony ostium created at the time of the primary DCR.

■During revision external surgery, the bony ostium is more easily enlarged through the endonasal approach.

■In revision cases a detailed examination should be performed to find the possible causes and the site of pathology.

■The rhinostomy from the sac into the lateral nasal wall should be created at the correct location and with the correct size, which should be over 5 mm wide.

■A low rhinostomy may not bypass a midsac or up- per-sac obstruction. A high rhinostomy leaves the nasolacrimal duct as a blind pouch that is not adequately drained. Despite a patent anastomosis, re-

tention produces sump syndrome.

■ A larger rhinostomy removing enough bone (15 mm × 15 mm) may prevent sump syndrome.

■Any nasal or paranasal abnormalities should be corrected during DCR surgery.

■If the rhinostomy is placed in a wrong location in the previous surgery, the new rhinostomy should be opened in the correct location.

■The success rate is directly related to the appropriate technique — inadequate surgical training and the lack of proper instrumentation decrease the success rate.

Introduction

Epiphora is one of the most prevalent functional disabilities of the ocular system. Various techniques (Table 27.1) have been developed to treat lacrimal diseases, to eliminate the infection, and to reconstruct a functional drainage pathway using very sophisticated instruments.

The practice of lacrimal surgery has been done by an external approach for a long time. Despite much debate, many ophthalmologists still believe that external DCR provides higher success rates than endoscopic DCR. The surgical treatment of DCR is very closely related to the inside of the nose. The problem of an endonasal approach was the difficulty in visualizing the endonasal anatomy. The introduction of microscopes and endoscopes into medicine opened the door for visualization of the interior of the nose, and endoscopic or microscopic endonasal

lacrimal surgery has become more popular in recent times [21]. Long-term success rates with endonasal techniques have been equivalent to that achieved with external DCR [10]. However, it is difficult to make evidence-based determinations about the efficacy of endonasal or external approaches. Each approach has its own advantages and disadvantages (Tables 27.2–27.5).

The endonasal approach is well suited for revision DCR surgery because the residual lacrimal sac can be accessed directly through the previous bony ostium created at the time of primary DCR, and the bony ostium is more easily enlarged [28]. The endoscopic DCR in revision cases also offers several other advantages including a complete exposure of those anatomical anomalies or inflammatory changes of the normal structures that commonly hinder conventional external surgery. Before revision surgery, a detailed preoperative evaluation is necessary.

Preoperative Evaluation

Although the tear drainage system appears very simple, draining the tears through the nasolacrimal system with

the help of gravity is indeed an intricate process. Drainage of tears depends on the volume of tear production, eyelid position, pump mechanisms, anatomy of the lacrimal system, gravity, and nasal air convection currents. Although clinical evaluation of gross lacrimal function is not difficult and can be made on the basis of history, determination of the cause may be extremely difficult and requires a variety of diagnostic procedures [7].

Lacrimal drainage dysfunction can be due to an anatomic obstruction, such as nasolacrimal duct fibrosis, or physiologic dysfunction from a failure of functional mechanisms (for example lacrimal pump inadequacy caused by poor orbicularis muscle tone); therefore, the diagnosis of the cause of epiphora is important. A list of tests required for the diagnosis of epiphora is given in Table 27.6.

The routine preoperative evaluation includes dacryoscintigraphy or dacryocystography. Dacryocystography is a safe, quick, and easy procedure using a radio-opaque material. It is widely established for the demonstration of stenosis. This procedure should not be performed in the presence of active dacryocystitis. Dacryocystography may be useful in demonstrating localized stricture, par-

tial obstruction, lacrimal diverticuli, fistulae, dacryoliths, and extrinsic and intrinsic tumors of the lacrimal drainage system [7].

The disadvantage of dacryocystography is that it provides restricted functional information, as in dysfunction of the canalicular muscle pump, slight narrowing of the ductal lumen and mucous membranes, since intubation of the canaliculi and active injection of the contrast material may overcome stenosis. Dacryoscintigraphy is also a simple, noninvasive physiologic test. Dacryoscintigra-

phy may provide information about physiological function. Limiting factors are the methodologically inherent minimal morphologic information and relatively large variations of normal transit times. Dacryocystography gives finer anatomic detail; however, dacryoscintigraphy is a more physiologic assessment since no instrumentation is necessary [7, 14]. In revision cases, functional dysfunction and canalicular obstruction should be considered in the differential diagnosis of epiphora before surgery.

238

Dye (fluorescein) disappearance test

Primary Jones dye test (Jones I and Jones II)

Lacrimal irrigation

Ultrasonography

Dacryocystography

Radionuclide dacryoscintigraphy

Computed tomography

Computed tomography dacryocystography

Magnetic resonance imaging

Magnetic resonance dacryocystography

Lacrimal endoscopy

Revision Dacryocystorhinostomy

Reasons for Failure

Failure of dacryocystorhinostomy (DCR) is attributable to a variety of causes (Table 27.7). The majority of cases were found to be related to internal nasal problems. A small and inappropriately placed bony opening, inadequate and improperly opened rhinostomy, excessive scar tissue production, anatomical abnormalities, and

Metin Onerci

concomitant paranasal sinus infections are the most important causes for the failure of DCR surgery.

Functional obstruction can be diagnosed with dacryoscintigraphy. If it is due to dysfunction of the lacripump system , the patient should be informed about prognosis. Narrowing of the nasolacrimal canaapparatus sometimes mimics the functional dys-

. However, these patients may benefit from the at the very least their symptoms improve after [41].

Approach

endoscopic or microscopic DCR is generpreference of many authors in instances of failed [8, 22, 28]. The angled instruments developed for sinus surgery allow the occluded ostium to

relatively safely evaluated under direct endoscopic

. Under direct endoscopic or microscopic visualization, the ostium can be enlarged and properly positioned to increase the likelihood of continued patency [19]. Granulations can be easily trimmed and cauterized, adhesions are cut together with resection of the anterior end of the middle turbinate. A small, high rhinostomy may cause sump syndrome. The sac should not be opened very high up without opening it inferiorly in order to prevent sump syndrome (Figure 27.1). Placement of the ostium too close to the middle turbinate results in subsequent adhesion and occlusion,. The endonasal approach is a one-stage procedure that permits correction of associated nasal disorders, such as septal deviation, middle-