Sir,

Periocular injections of steroids are commonly used in the treatment of posterior segment intraocular inflammation. We present the case of a localised abscess following subtenon injection of triamcinolone.

Case report

A 33-year-old woman presented with a 10-day history of floaters, blurred vision, and a painful red left eye. She was systemically well. Past history included extracapsular cataract extractions and lens implants for secondary cataract (following previous long-term systemic steroid use for mononeuritis multiplex).

Visual acuities (VA) were 6/5 and 6/60 in the right and left eye. There was no relative afferent pupillary defect and a mild anterior uveitis and marked vitritis in the left eye with associated cystoid macular oedema (CMO) was noted. Intraocular pressures were normal. There was no associated retinal vasculitis or optic disc abnormalities. Examination of the right eye was unremarkable. All investigations including inflammatory markers, full blood count, renal function tests, serum-ACE, autoimmune profile, chest X-ray, and previous MRI were normal. A diagnosis of intermediate uveitis was made and she was treated with topical dexamethasone 0.1% followed by a posterior subtenons injection of 40 mg/1 ml of triamcinolone acetonide (Kenalog, Bristol Myers Squibb). Under topical anaesthesia, a short 25G needle was passed through the bulbar conjunctiva as far posterior as directly visible aiming for the subtenon space in the superotemporal fornix with the patient looking inferonasally. The tip was swept from left to right during advancement, and the limbus observed for any movement to ensure that the needle remained superficial to the sclera (technique described by Tanner et al1). A single drop of chloramphenicol was administered following the injection. The procedure was uneventful, and the vitritis, CMO, and VA improved to 6/12 at 1 week.

She returned 2 weeks after the injection with pain, redness, and yellow discharge in the left eye. There was no evidence of intraocular inflammation and a localised area of hyperaemia and sub-conjunctival infiltrate was noted at the site of the subtenon injection (Figure 1). Culture yielded Streptococcus pneumonia and she was started on oral and topical ciprofloxacin (guided by culture sensitivities). Systemic ciprofloxacin was stopped and topical treatment tapered after 10-days following improvement in symptoms and infiltrate. Within 2 days of stopping the oral medication she noticed increased discomfort of the left eye, but there was no visible subconjunctival infiltrate or intraocular inflammation. She was restarted on oral ciprofloxacin and the topical treatment continued. Her symptoms and signs rapidly resolved within 1 week and only a deposit of steroid supernatant remained visible at the injection site (Figure 2).

Figure 1
figure 1

Subconjunctival abscess in the superotemporal area.

Full size image
Figure 2
figure 2

Resolution of abscess after 1 week of treatment. Note that the deposit of steroid supernatant is still visible.

Full size image

Three months after initial presentation she was asymptomatic, VA was 6/5 bilaterally and there was no residual signs of infection, intraocular inflammation, or CMO.

Comment

Treatment options for intermediate uveitis include local or systemic immunosuppresion.2, 3 Local injection into the orbital floor or subtenon space has the advantage of facilitating high local concentration of steroid to the posterior segment while minimising unwanted systemic effects, and has been shown to be safe and efficacious in the treatment of posterior uveitis.1, 4

Reported complications include raised intraocular pressure, inadvertent intravascular injection, globe perforation, cataract, allergy, strabismus, fat atrophy, and conjunctival necrosis.1, 4, 5, 6, 7

Engelman et al8 previously reported an orbital abscess following posterior subtenon injection of triamcinolone. The patient presented 3 weeks after a second injection with an orbital mass, proptosis, and blepharoptosis, which was treated surgically by drainage and with systemic antibiotics. In our case the small, localised abscess was confined to the site of the subtenon injection, required no drainage, and resolved on oral and topical antibiotics.

The current report while illustrating the efficacy of local steroid in the treatment of intermediate uveitis highlights a further late potential complication of subtenons injection. Further studies comparing the safety and efficacy of different routes of periocular steroid administration and routine use of povidine iodine preoperatively and more prolonged course of topical antibiotics following subtenons injection may be warranted to assess and minimise the risk of this rare complication.