February 01, 2014
7 min read
Save

Eyelid swelling and itching occur after tube shunt implantation

The elderly patient received a Baerveldt tube shunt 4 days before presentation and was concerned about possible allergies to postop medications.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

An 84-year-old woman who 4 days earlier had a Baerveldt tube shunt implanted in the left eye for primary open-angle glaucoma called the on-call ophthalmologist with complaints of left eyelid swelling, itching and tearing. She expressed concern about possible allergy to her postoperative medications. She denied any feeling of pain but described a “mild tugging feeling.” She stated she was unable to assess any vision changes due to her eyelid swelling. The patient was immediately seen that day.

Her medical history included hypertension, coronary artery disease and hypothyroidism, with current systemic medications of levothyroxine and amlodipine. Her ocular medications included Travatan (travoprost, Alcon) in both eyes at bedtime and postoperative Maxitrol (neomycin, polymyxin B, dexamethasone, Alcon) four times a day in the left eye. She had an extensive medication allergy list, which included Cosopt (dorzolamide hydrochloride-timolol maleate ophthalmic solution, Merck), Xalatan (latanoprost, Pfizer), Betoptic (betaxolol ophthalmic, Alcon), Lipitor (atorvastatin, Pfizer), metoprolol, omeprazole, Plavix (clopidogrel, Bristol-Myers Squibb), Procardia (nifedipine, Pfizer), Propine (dipivefrin, Allergan) and prednisone. None of these allergies had a documented reaction. Her previous ocular surgeries included cataract extraction with posterior chamber IOL placement in both eyes and the aforementioned filtering surgery in the left eye.

Examination

On initial presentation, the patient had an uncorrected visual acuity of 20/60+ in the right eye and 20/80– in the left eye. Pupil exam showed sluggish but reactive pupils in both eyes. Extraocular movements were full on initial assessment, and external exam showed 2+ upper lid, lower lid and periorbital edema with 1+ associated erythema in the left eye. There was no significant warmth on palpation. IOP was 21 mm Hg in the right eye and 26 mm Hg in the left eye. Anterior exam was normal in the right eye. In the left eye, there was 2+ diffuse chemosis and injection, but the tube was in a good position and not exposed. The cornea showed 1 to 2+ edema with Descemet’s folding, and the anterior chamber had only trace cell. The posterior exam showed glaucomatous changes in the optic nerves of both eyes, clear vitreous in both eyes, and a normal-appearing retina and macula in both eyes.

Approximately 45 minutes after the initial exam, the patient began to complain of increasing periorbital pain. Confrontation visual fields and color vision were found to be normal. Hertel was unable to be obtained due to the patient being uncooperative secondary to pain but appeared to clinically be proptotic. The extraocular muscles were reassessed and found to be slightly limited in the medial and lateral directions.

She was sent for basic lab workup, which had no leukocytosis, and an orbital CT scan, which showed soft tissue swelling and intraorbital fat stranding (Figures 1 and 2). After starting IV vancomycin and Unasyn (ampicillin and sulbactam, Pfizer), a Hertel exam measured 16 mm in the right eye and 23 mm in the left eye, with a base of 107 mm. A repeat extraocular muscle exam approximately 4 hours from initial presentation showed severe movement limitations in all directions of gaze (Figures 3 and 4).

Figure 1. 
Figure 2. 

Figures 1 and 2. Orbital CT scan with contrast, showing periocular soft tissue edema, enhancing Ahmed implant and intraorbital fat stranding.

Images: Hansen BA, Krishnan C

Figure 3.

Figure 3. Patient approximately 2 hours after initial presentation.

Figure 4.

Figure 4. Patient approximately 4 hours after initial presentation.

PAGE BREAK

What is your diagnosis?

Tube shunt implantation

Orbital cellulitis should be suspected in any patient who presents with orbital pain, proptosis, limitation of ocular motility, lid edema or orbital congestion, and has periorbital cellulitis or a bacterial infection within the paranasal sinuses, dental region or from ocular implants. Conditions to consider in the differential diagnosis are orbital pseudotumor, orbital tumor, acute allergic reaction, dacryocystitis and dacryoadenitis. Orbital pseudotumor can mimic orbital cellulitis; however, this is a diagnosis of exclusion, and infectious cause must be ruled out. In light of the recent tube shunt placement and surgical intervention in our patient, orbital cellulitis was high on the differential.

Due to the rapid advancement of the condition, a decision was made to remove the implant emergently. The device was removed without complication. It was noted intraoperatively that there was mucopurulent discharge around the plate at the time of removal. Within 24 hours, cultures of the implant grew methicillin-sensitive Staphylococcus aureus.

Discussion

Historically, postoperative complications of drainage device placements have been divided into early or late categories. Early complications include bleeding, retinal detachment, aqueous misdirection, endophthalmitis, hypotony, inflammation and tube blockage. Late complications include hypotony, increased IOP, tube migration, implant erosion, corneal decompensation, cataract, calcification of the implant, endophthalmitis and strabismus. Orbital complications are an uncommon phenomenon. Documented cases have included orbital cellulitis, myositis, significant proptosis and silicone oil migration. Orbital cellulitis, in particular, is a rare postoperative complication after tube shunt implantation. To our knowledge, only six previous cases have been reported.

Orbital cellulitis is a potentially vision- and life-threatening condition, making early diagnosis essential. The risk of infectious spread is thought to be secondary to the relatively large potential space within the orbit that is drained by the superior and inferior orbital veins (of which the inferior are valveless), which communicate directly with the cavernous sinus. The diagnosis is more common in children, with 86% to 98% of cases having coexisting rhinosinusitis as the likely etiology of infection. Other documented causes include ophthalmic surgery (strabismus repair and blepharoplasty being most common), peribulbar anesthesia, penetrating trauma, dacryocystitis, teeth/middle ear/facial infections and infected mucocele with orbital erosion.

Orbital cellulitis is primarily a clinical diagnosis with important distinguishing features of pain with ocular movements, proptosis and ophthalmoplegia. Both orbital and preseptal cellulitis may cause ocular pain, eyelid swelling and chemosis. Fever and leukocytosis also have a low sensitivity and specificity in aiding diagnosis. The diagnostic features on initial exam of our patient were not entirely clear with a question of possible allergy, given her extensive documented allergies, vs. normal postoperative changes. Interestingly, two cases of orbital cellulitis reported by Marcet et al after Ahmed shunt implantation also presented with an initial mixed clinical picture and question of possible allergy. These patients presented on postoperative day 4 and day 2 with proptosis, eyelid swelling and mild limited ocular motility. One patient had significant pain, and the other patient had no significant erythema.

Although orbital cellulitis is primarily a clinical diagnosis, CT imaging is typically used as a confirmatory test or to rule out abscesses or other complications. However, the utility of postoperative CT, as in this case, to clarify the diagnosis is unclear. In these patients, it may be difficult to decipher between normal postoperative enhancement and early infection. This is particularly true when the patient presents very soon postoperatively. Of the six published cases of orbital cellulitis after drainage device surgery, only three had CT scans that definitively confirmed the diagnosis. Supplemental testing such as complete blood count, metabolic panel, blood cultures (if febrile) and spinal tap (if meningitis symptoms are present) can also aid in decisions for treatment and prognosis.

PAGE BREAK

Treatment

Treatment for orbital cellulitis should be directed toward likely infectious organisms, with the most common being Staphylococcus and Streptococcus species. Alpha and beta hemolytic strep are most common (17% to 54%), followed by methicillin-sensitive S. aureus (12% to 24%), methicillin-resistant S. aureus (1% to 20%) and more rare causes, including Haemophilus influenzae, Aeromonas hydrophila, Pseudomonas aeruginosa, Eikenella corrodens, mucormycosis and aspergillus. General recommendations include IV antibiotics until ocular symptoms resolve; a switch to oral antibiotics can be made if the patient is afebrile with no abscesses or extension of the infection for a total of 2 to 4 weeks.

The additional question concerning treatment in a postoperative setting is whether or not to remove surgical hardware. Three of the six published cases described patients on postoperative months 1, 8 and 15 who were diagnosed with tube exposure after conjunctival erosion. In these cases, the exposure was thought to have been the nidus for the orbital infection, and the implants were removed. The other published cases, presenting on postoperative days 9, 4 and 2 without tube exposure, were treated with IV antibiotics alone.

Our patient showed remarkably rapid disease progression, moving from unrecognizable motility changes to severe restriction within approximately 4 hours of presentation despite initiation of IV antibiotics. The reason for her rapid progression was not entirely clear. In general, the initial bacterial load, virulence of the organism, properties of the affected tissue, and whether or not there is a continued nidus can explain the severity of the infections. Based on our patient’s progressive course, the decision was made to remove the implant as a potential source of infection. This decision should be made on an individual basis, taking in to account the visual prognosis and potential.

Follow-up

Our patient had a 7-day inpatient stay and showed rapid improvement after implant removal and daily IV antibiotics. After the cultures returned, vancomycin and Unasyn were discontinued, and she was started on IV oxacillin. She was discharged to a rehab facility and received a total of 14 days of IV antibiotics. Her last available follow-up appointment on postoperative week 4 showed a vision of 20/200 in the left eye. Her IOP in the left eye was slowly increasing, with a most recent pressure of 30 mm Hg, with plans for a possible repeat drainage operation in the future.

References:
Botting AM, et al. Int J Pediatr Otorhinolaryngol. 2008;doi:10.1016/j.ijporl.2007.11.013.
Chaudhry IA, et al. Eur J Ophthalmol. 2007;17(1):136-140.
Danesh-Meyer HV, et al. Arch Ophthalmol. 2002;120(6):846-847.
Durand, ML. Periocular infections. In: Bennett, et al, eds. Principles and Practice of Infectious Diseases. 7th ed. Philadelphia: Churchill Livingstone Elsevier; 2010:1569.
Guerrero AH, et al. Int Ophthalmol Clin. 2000;40(1):149-163.
Hofbauer JD, et al. Am J Ophthalmol. 1994;118(3):391-392.
Karr DJ, et al. J Pediatr Ophthalmol Strabismus. 1990;27(2):107-110.
Laviña AM, et al. Arch Ophthalmol. 2002;120(6):849-851.
Marcet MM, et al. Ophthal Plast Reconstr Surg. 2005;doi:10.1097/01.IOP.0000150351.91768.16.
Oh KT, et al. Am J Ophthalmol. 1999;128(3):375-376.
Weakley DR. Ann Ophthalmol. 1991;23(12):454-457.
For more information:
Bradley A. Hansen, MD, and Chandru Krishnan, MD, can be reached at New England Eye Center, Tufts University School of Medicine, 750 Washington St., Box 450, Boston, MA 02111; 617-636-4219; fax: 617-636-4866; website: www.neec.com.
Edited by Jennifer Renz, MD, and Avneet K. Sodhi, MD. They can be reached at New England Eye Center, Tufts University School of Medicine, 750 Washington St., Box 450, Boston, MA 02111; 617-636-4219; fax: 617-636-4866; website: www.neec.com.