Endoscopic Subperiosteal Abscess Drainage

Right Orbital Subperiosteal Abscess Drainage

Acute rhinosinusitis (ARS) accounts for one-fifth of all adult and pediatric antibiotic prescriptions. Bacterial ARS can lead to orbital or intracranial infections by direct or hematogenous spread. If left untreated, this can result in permanent vision loss, meningitis, intracranial abscess, sepsis, and death. The incidence of serious complications from ARS has been estimated to be 1:12,000 in children and 1:32,000 in adults. Orbital complications are more common than intracranial complications and most commonly occur in male children.

Orbital complications have historically been categorized by Chandler’s classification, as shown in Table 30.1 . Most orbital complications occur from an infected ethmoid sinus, whereas the other paranasal sinuses are less frequently the source of the infection. Theories for explaining orbital spread include congenital dehiscence of the lamina papyracea, direct spread via ethmoid artery foramina, and the presence of valveless venous anastomoses draining the ethmoid and maxillary sinuses.

Table 30.1

Chandler Classification of Orbital Complications of Sinusitis

Grade Symptoms
1 Preseptal cellulitis
2 Orbital cellulitis
3 Subperiosteal abscess
4 Orbital abscess
5 Cavernous sinus thrombosis

Subperiosteal abscesses (SPAs) of the orbit most commonly affect the medial wall but can also involve the inferior and superior orbital walls. In the past, surgical approaches for abscess drainage involved open orbitotomies and external approaches to the sinuses. More recently, endoscopic sinus surgery has largely replaced open techniques, as this obviates the need for a facial incision. However, visualization can be difficult owing to bleeding from the inflamed mucosa. This chapter illustrates the surgical management of SPAs that are amenable to transnasal endoscopic drainage.

Clinical Presentation

Children are frequently seen in clinics with symptoms of fever, nasal congestion, nasal drainage, and facial pain consistent with ARS. Those with orbital complications are distinguished by ophthalmologic symptoms including eye swelling, blurry vision, pain, and limited ocular movements. Examination can demonstrate upper and lower eyelid swelling, decreased visual acuity, ophthalmoplegia, chemosis, and/or proptosis. SPAs may be difficult to distinguish from orbital cellulitis by clinical examination, but lateral or inferior displacement of the globe is suggestive of abscess formation. Nasal endoscopy may show swollen turbinates and purulent nasal drainage.

The initial workup includes a full ophthalmologic evaluation including assessment of the pupil, retina, intraocular pressure (IOP), extent of proptosis, and presence of chemosis. Laboratory tests, including complete blood count, basic metabolic panel, and inflammatory markers, are useful to establish baseline values and trends in cases that are observed or do not quickly resolve. If there is a concern for an orbital complication beyond preseptal cellulitis, computed tomography (CT) of the sinuses, preferably with contrast, is the investigation of choice. It is useful for preoperative planning to determine the extent of infection and specifically to exclude involvement of the cavernous sinus. A CT scan is quick and readily available, does not normally require sedation, and defines the bony and soft-tissue anatomy well. Many authors advocate broad-spectrum intravenous antibiotics for 24 to 48 hours and a CT scan only if there is worsening or no improvement. If there is concern for intracranial complications or invasive fungal sinusitis, magnetic resonance imaging (MRI) should be performed. However, MRI is not indicated for routine workup for orbital complications of ARS. Figs. 30.1 and 30.2 illustrate examples of SPAs.

Fig. 30.1

Computed tomography scan of sinus with contrast, coronal cut showing left subperiosteal medial and superior abscesses ( arrow ).

Fig. 30.2

Computed tomography of sinus with contrast, axial cut showing a right medial subperiosteal abscess ( arrow ).


Medical management should be initiated in all children with SPAs and includes antibiotics and nasal hygiene, including high-volume saline rinses and a short course of decongestants (i.e., oxymetazoline). Initial antibiotic therapy may be empiric and cover the most common responsible organisms. Coudert et al. reviewed 48 children with SPAs and found that 60% of cultured abscesses grew Streptococcus, 12% Staphylococcus , and 12% anaerobic species. We recommend starting a regimen of a single-agent antibiotic such as ampicillin-sulbactam or a third-generation cephalosporin such as ceftriaxone in children 9 years or younger, in whom polymicrobrial infections is less common. If the child has a penicillin allergy, clindamycin is used instead. Liao et al. reported a 6.5% (3/46) rate of methicillin-resistant Staphylococcus aureus (MRSA) on culture and recommended that the initial broad-spectrum antibiotic regimen should include MRSA coverage. Based on our hospital antibiogram, we do not need to use empiric antibiotic therapy that includes MRSA coverage, but the need to do so may differ in other regions. Conversely, in a child 10 to 15 years or older in whom the infection has a higher chance of being polymicrobial or odontogenic in origin, broader anaerobic and MRSA coverage is part of initial antibiotic therapy. If there is concern for meningeal involvement or associated intracranial complications, double coverage with a third-generation cephalosporin and metronidazole is indicated at a dose appropriate to pass through the blood–brain barrier

The decision to start medical treatment and observe versus drainage of a SPA is a critical part of the decision making. A neurologic and ophthalmologic examination is crucial. Patients with a normal visual acuity, pupil, and retina and with no ophthalmoplegia, with an IOP less than 20 mm Hg and proptosis less than 5 mm, may be treated with medical management and close observation. Patients with compromised vision or rapid progression to intracranial complication need immediate drainage. “Immediate drainage” is not well defined, but in our institution it is within 12 to 24 hours of presentation.

Age plays a determining factor in the decision making to drain a SPA. Children older than the first decade of life tend to have more aggressive bacteria by culture with a higher likelihood of anaerobes. Older children and adults are more prone to intracranial complications of sinusitis. In any older child or adult who presents with an orbital complication of sinusitis, there should be no hesitation to proceed with surgical decompression.

Several studies have investigated abscess width and volume on CT as factors considered for initial medical versus surgical management. Abscess width of less than 10 mm with normal findings on ophthalmologic examination may be medically treated initially. Abscess volume greater than 500 mm 3 usually requires surgical intervention. These measurements may be difficult to determine consistently, as there is no standard way to measure width or volume on a CT scan.

Thus in children younger than 10 years without visual compromise, normal IOPs, smaller abscess size, and no neurologic involvement, we recommend medical management with close observation. It is worth noting that children with a large-volume SPA (volume > 500 mm 3 ) may have a longer hospital stay and duration of antibiotic therapy and a higher incidence of peripherally inserted central catheters. Patients whose conditions do no improve after 48 to 72 hours or in whom worsening occurs in 48 hours should be considered for surgical management; decision making requires close communication between the otolaryngology and ophthalmology services. We do not recommend routine repeat imaging before surgery, but studies on this subject are lacking.

Surgical Management

The location of the SPA directs the approach. Most medial and inferior abscesses are amenable to endoscopic drainage. Superior and lateral abscesses are less common and less accessible endonasally and are more likely to require an external orbitotomy. Endoscopic sinus surgery for acute SPAs is challenging, primarily owing to the inflamed and hyperemic mucosa that can lead to increased blood loss and poor visualization (see Video). The setup is similar to functional endoscopic sinus surgery, and intraoperative image-guided navigation is recommended. The efficacy of reconstructed non–image-guided CT scans used for endoscopic surgery is unknown, although we would rarely recommend a repeat scan for the purpose of using image-guided surgery.

Noninvasive measures for hemostasis should be used, including bed elevation, keeping the blood pressure low, and topical vasoconstriction with oxymetazoline pledgets. The nasal mucosa is injected with 1% lidocaine with 1:100,000 epinephrine. Fakhri describes injections of the lateral nasal wall, using as few injections as possible to minimize bleeding from these sites. We recommend injecting the head and axilla of the middle turbinate, and, if able to visualize, the region adjacent to the sphenopalatine artery posterior to the maxillary sinus along the lateral nasal wall.

Traditionally a 4-mm endoscope is used; however, in small children, a 2.7-mm endoscope may be necessary. A smaller-diameter scope does compromise visualization but may be the best option in a small swollen nasal cavity. The middle turbinate is medialized, and an uncinectomy is performed and the maxillary sinus ostium identified. Using a 30-degree endoscope, a maxillary antrostomy is performed to drain the maxillary sinus and to provide a surgical landmark for the orbital floor. Purulence should be collected for culture when encountered. The ethmoid bulla is then entered and an anterior ethmoidectomy is performed, exposing the middle turbinate lateral to the lamina papyracea. Based on the size and posterior extent of the abscess, the basal lamella of the middle turbinate is entered and a posterior ethmoidectomy is performed to provide posterior access to the medial orbit. A posterior-to-anterior total ethmoidectomy can then be performed along the skull base, taking care to fully expose the lamina laterally. Froehlich et al. report good results with a limited anterior ethmoidectomy with dissection immediately lateral to the lamina in a small number of children with SPAs. However, the authors continue to recommend a more extensive dissection because we feel it is less likely to lead to incomplete drainage.

The lamina is inspected for dehiscences and spontaneous purulent drainage. If no areas of purulence are noted, the lamina is entered sharply with either a curette or a periosteal elevator and the abscess is drained with assistance of external pressure on the orbit. The lamina does not need to be completely removed after the abscess is drained. Good hemostasis is achieved and nasal packing is avoided, maximizing decompression. If intraoperative bleeding obscures safe access to the orbit, an external approach should be used (see Video).

Further research is indicated for management of the uninvolved side, as well as the utility of concurrent adenoidectomy in younger patients. Although most medially based abscesses along the lamina papyracea are amenable to endonasal endoscopic drainage, the presence of a superolateral component may require a combined procedure with an ophthalmologist for an orbitotomy ( Fig. 30.3 ). The presence of an intraconal abscess requires intraoperative ophthalmologic input and may require serial measurements of IOP.

Jan 3, 2021 | Posted by in OPHTHALMOLOGY | Comments Off on Endoscopic Subperiosteal Abscess Drainage

Full access? Get Clinical Tree

Get Clinical Tree app for offline access