Pediatric Plot Twist

A healthy 9-year-old girl presented to an urgent care clinic with a red, tender, and swollen right upper eyelid. She was prescribed a 10-day course of hydroxyzine and cephalexin. The family sought care with another eye care provider (not me) approximately 10 days into the course of antibiotics because the edema had not resolved. That eye care provider advised the family that the girl should complete the course of treatment and return if no improvement was noted after 1 additional week. The patient did not return for care until approximately 1 month later, when she presented to my clinic with right upper lid swelling that had now been present for approximately 2 months.

INITIAL PRESENTATION

The patient reported no pain or double vision, and she was afebrile. Her medical history was unremarkable, and she had no known chronic or recurrent ocular problems. UCVA measured 20/20 OU. Pupils were equal, round, and reactive to light in bright conditions (3.0 mm in the right eye and 3.0 mm in the left). However, anisocoria was noted in the dark (4.5 mm in the right eye and 5.5 mm in the left), in addition to a possible dilation lag OD. No relative afferent pupillary defect was observed. Extraocular muscle movements were full and without notable restriction in both eyes.

External examination of the right eye revealed moderate ptosis of the right upper lid with subtle serous edema (Figure 1). Further lid evaluation showed that levator function was reduced to 9 mm in the right eye, compared to 14 mm in the left. All other anterior and posterior segment right eye findings were within normal limits. Examination of the anterior and posterior segment of the left eye was unremarkable.

Photos of the patient’s affected eye on her grandmother’s mobile phone, taken at the time of her presentation to the urgent care clinic, suggested that the condition had resembled either a preseptal cellulitis or allergic blepharoconjunctivitis.

FURTHER INVESTIGATION

My primary concern for this patient was that she had a subacute Horner syndrome. However, her significantly reduced levator function did not align well with that diagnosis. Was this an incidental finding caused by the presence of subtle edema, or did this localize the lesion to the orbit?

The child and family denied anhidrosis, and I was able to confirm the presence of bilateral hidrosis and flushing by asking the child to walk outdoors for an extended period and then return to the exam room. There was no history of chest or neck surgery.

Given the pupil findings, a shared decision was made with the family to urgently evaluate possible etiologies via laboratory studies and neuroimaging. This was necessary to avoid delaying a potentially fatal diagnosis, as my clinic did not have cocaine or apraclonidine drops immediately available to formally confirm Horner syndrome.

An MRI of the head, neck, and chest with and without contrast was ordered and obtained within 2 days. A complete blood count was obtained to evaluate for infection and hematologic cancers. Because neuroblastoma has been implicated as a common cause of acquired pediatric Horner syndrome, a 24-hour urine catecholamine test was also ordered.¹ However, the sample was not returned and therefore the result was never obtained.

RESULTS

The results of the complete blood count, MRI of the neck, and MRI of the chest were noncontributory. However, the MRI of the head was alarming. Neuroimaging confirmed the presence of a mass in the posterosuperior right orbit with suspected involvement of the superior rectus and superior oblique muscles (Figures 2 and 3). Complex signal characteristics were noted in addition to the appearance of bony expansion or destruction of the medial posterior orbit.

According to the radiology report, this was most concerning for aggressive etiologies such as rhabdomyosarcoma or venous lymphatic malformations, and it was felt that inflammatory etiologies were unlikely due to the lack of retrobulbar fat edema. A CT scan of the orbits was recommended to further evaluate characteristics of the mass, if clinically necessary.

THE CARE TEAM

Management plans were put in place before calling the mother with the MRI results. The child’s pediatrician was immediately contacted to discuss the findings and the urgent need for elevation of care. It was apparent that the child would need a surgical biopsy and involvement of pediatric oncology. The pediatrician was immensely helpful in coordinating care, thanks to her close working relationship with a large pediatric hospital in the area. I drafted a referral letter summarizing my examination findings and forwarded it to pediatric oncology, along with the MRI results. The patient’s mother obtained a copy of the MRI images on a disc to carry by hand to the hospital.

The patient reported to the pediatric emergency department the following day and was expeditiously admitted to the oncology ward for initial evaluation. The attending oncologist consulted a pediatric craniofacial surgeon and a pediatric general surgeon, and collectively they formulated an extensive care plan. The patient was taken to the OR, and under anesthesia a bone marrow biopsy, a biopsy of the orbital mass, and placement of a central venous port was completed in anticipation of the need for chemotherapy.

PLOT TWIST

Shortly after this procedure was completed, the pathology results were obtained. The biopsy results showed presence of inflammatory cells consistent with a diagnosis of orbital inflammatory myositis. No cancerous cells were identified on the specimens, and aggressive etiologies were excluded. As a result, the central venous port was immediately removed.

Upon confirmation of the diagnosis, the patient was immediately started on 20 mg oral prednisone daily, and her condition rapidly improved. Further bloodwork and chest x-rays were ordered to screen for sarcoidosis, granulomatosis with polyangiitis, and other autoimmune disorders, which later came back negative. She was discharged on oral prednisone under the direction of the oncologist a few days later.

The patient’s condition resolved after approximately 2 weeks, and a taper of the oral prednisone was initiated. Surprisingly, her condition recurred about 1 month after the prednisone taper, so she was placed on mycophenolate mofetil (CellCept, Genentech) for chronic immunosuppression. She continued the mycophenolate for 6 months after the initial recurrence and was subsequently tapered off the medication without additional recurrences.

REVIEWING THE LITERATURE

Nonspecific orbital inflammation (NOI), previously referred to as orbital pseudotumor, is an umbrella term for inflammatory masses of the orbit. Although NOI is the third most common orbital mass in adults, a 2008 review article of reports found only 68 cases of pediatric NOI dating back to the 1950s.^2,3 The most common presenting symptoms in both adults and children include pain, diplopia, extraocular muscle restriction, and proptosis. In one case series, 21% of pediatric cases were reported to have had pupillary involvement.⁴

Diffuse or localized NOI presentations have been reported with inflammation possibly affecting the lacrimal gland, extraocular muscles (as in this case), uvea, and/or the cavernous sinus.¹ Although adult NOI most often presents unilaterally, pediatric NOI frequently presents bilaterally with additional presence of uveitis and optic disc edema.³ Because of the variable presentation and overlap with other disease entities, utilization of neuroimaging modalities such as CT or MRI of the orbits is necessary to exclude malignant, infectious, and other etiologies. Biopsy is indicated for atypical cases prior to initiation of immunosuppressant medications.⁵

The pathogenesis of NOI is not well understood, but the disease has been associated with various autoimmune conditions, including systemic lupus erythematosus, thyroid disease, sarcoidosis, rheumatoid arthritis, Crohn disease, myasthenia gravis, ankylosing spondylitis, and others. Recently, immunoglobulin G4–related disease has been implicated in 17% to 60% of adult cases of NOI as an etiology, but incidence and prevalence in the pediatric population has not been widely studied.^6,7 Because of these systemic associations, a systemic workup is indicated for suspected NOI in either adults or children.

Various therapies have been successful for NOI in adults and children. Oral corticosteroids are generally the first-line therapy, but use of NSAIDs, systemic immunomodulatory medications, radiation therapy, and surgical resection has also been suggested, especially for chronic or recalcitrant disease.^8,9

KEY TAKEAWAYS

In daily clinical practice, we often see patients with atypical clinical features or with signs and symptoms that do not follow the expected clinical course. Sometimes explaining the constellation of clinical findings in difficult cases can be challenging. These cases teach us to critically analyze our findings and data, widen our differential diagnoses, and involve other medical specialists when the waters become too murky.

Although the need for eye care providers to order neuroimaging for pediatric patients does not arise frequently, cases such as this one illustrate both the value and drawbacks of neuroimaging in making a definitive diagnosis. The case also demonstrates how a biopsy can drastically alter both the course of treatment and prognosis.