Orbital cellulitis in a cancer patient after a sinus surgery. A 69-year-old woman presented to our service with swelling, decreased visual acuity, and pain in the right eye. She had a history of a recurrent atypical pituitary macroadenoma with extension into the sinuses and CREST (calcinosis, Raynaud phenomenon, esophageal dysmotility, sclerodactyly, and telangiectasia) syndrome. She had undergone multiple surgical resections and received a total of 70 Gy of radiation. On examination , she was afebrile, the visual acuity was 20/60 in the right eye (compared with 20/25 at baseline), and there was a positive afferent pupillary defect. The patient had right eye proptosis, upper eyelid ptosis, restriction in movements, and pain in all directions of gaze (a, b). Ocular examination revealed conjunctival chemosis; there were no other intraocular findings. A laboratory work-up revealed leukocytosis (leukocyte count, 17.6 × 109/L) and an absolute neutrophil count of 14.2 × 109/L. Computed tomography showed right eye proptosis, periocular soft tissue thickening (c, arrow), intraorbital fat stranding (d, asterisk), enlargement of the extraocular muscles (d, arrow), and a bony defect in the medial wall from a previous sinus surgery (d). The patient was treated with broad-spectrum systemic antibiotics and antifungals, and the orbital cellulitis gradually resolved over 13 days of treatment
Orbital cellulitis in a cancer patient treated with chemotherapy. A 56-year-old woman with lung adenocarcinoma treated with local resection, radiation therapy, and chemotherapy presented to our service with worsening left eye irritation and progressive swelling during chemotherapy. On examination , she was afebrile, the best corrected visual acuity was 20/40 (compared with 20/25 at baseline), and the patient had full color vision and no afferent pupillary defect. The patient had left eye periocular edema and erythema with upper eyelid ptosis. She had left eye hypoglobus and proptosis of 2 mm compared to the right eye. Prominent conjunctival chemosis was noted with conjunctival exposure on eyelid closure (a). Extraocular movements were restricted in all directions of gaze; no further findings were noted on dilated fundus examination. A laboratory work-up demonstrated white blood cell count within normal limits. Computed tomography showed left eye proptosis and hypoglobus, periocular soft tissue thickening, induration and stranding of the intraconal fat (asterisk), and abnormal enhancement of the extraocular muscles (b). The patient was treated with broad-spectrum antibiotics, and clinical improvement was noted after 6 days
We also routinely obtain a consultation from an ear, nose, and throat specialist to make sure that nasal endoscopy is performed to rule out obvious fungal sinusitis.
Clinical Findings in Cancer Patients with Orbital Cellulitis
The clinical examination findings in cancer patients with orbital cellulitis may be similar to the findings in patients with orbital cellulitis in the general population and often include fever, periorbital soft tissue edema, erythema, warmth, tenderness, upper eyelid ptosis, chemosis, limitation of and/or pain with extraocular motility, and proptosis (Figs. 13.1a, b and 13.2a). Deeper orbital soft tissue involvement (postseptal involvement) may be accompanied by gradual visual loss, but in the absence of an orbital abscess, acute optic nerve compromise is unlikely. Since neutropenia is common in cancer patients, inflammatory signs in cancer patients with orbital cellulitis may be less severe than inflammatory signs in immunocompetent patients with orbital cellulitis.
The physician should assess functionality of all orbital structures, including the optic nerve (by testing for visual acuity, pupillary reaction, color vision, and visual field); motor cranial nerves III, IV, and VI (by testing for limitation in ductions); and sensory cranial nerves V1 and V2 (by testing for hypoesthesia of the cornea and periocular skin). A full ophthalmic examination including dilated fundoscopy should be performed with special attention to the optic disk and engorgement of venous vessels. Neurological signs and symptoms, such as headache, meningismus, multiple cranial neuropathies, and progression to involve the contralateral side, seizures, focal neurological deficits, and altered mental status, should alert the clinician to the possibility of intracranial involvement. In cancer patients, all these signs and symptoms should be sought. Further work-up and close monitoring are recommended even in patients with only preseptal signs without evidence of orbital involvement because patients with cancer and orbital cellulitis are prone to more rapid deterioration than is observed in patients without cancer.
Progression of Orbital Cellulitis to Orbital Abscess
The frequency of abscess in patients with orbital cellulitis is a subject of debate. One paper in the literature proposes immunocompromise as a risk factor for progression of paranasal or orbital cellulitis to the development of an abscess; however, the paper provides no data to support the claim . Interestingly, in a review by the senior author (B.E.) of radiographic studies in all cancer patients treated for orbital cellulitis at our tertiary cancer center over the past 10 years, no classic subperiosteal orbital abscess was found (unpublished data). The rarity of progression of orbital cellulitis to orbital abscess may be due to an alteration in the normal function of the immune system and lack of a robust inflammatory response to infectious agents in cancer patients. One study of the milieu of cytokines expressed in children with orbital cellulitis with a subperiosteal abscess revealed elevation in interleukin-1 (IL-1), IL-1 receptor antagonist, IL-6, and tumor necrosis factor as significant components . In cancer patients, changes in cellular physiology due to immune suppression may affect the expression of these inflammatory mediators and other cytokines, and this phenomenon may explain the rarity of orbital abscesses in the setting of orbital cellulitis in cancer patients.
In a 2015 review from the United Kingdom of 54 non-cancer patients with a diagnosis of orbital cellulitis seen between January of 2008 through December of 2012, 11 patients (20%) had subperiosteal abscesses, and an additional 5 patients (9%) had an orbital abscess . A retrospective review published in early 2015 showed that an abscess of any kind developed in 32 of 83 children (39%) with orbital cellulitis . In contrast, in the reports of orbital cellulitis in immunocompromised patients published thus far, no patients were found to have subperiosteal abscesses. Careful review of the literature reveals rare case reports of soft tissue abscess formation in the orbital and ocular adnexal tissues in immunocompromised patients, but none of these abscesses are subperiosteal abscesses [25, 26]. In a 2004 case series of seven immunocompromised patients with fungal orbital cellulitis, no abscesses were noted; however, nearly all patients had extensive opacification of the sinuses with varying degrees of bony destruction . In 18 years of practice at our tertiary cancer center, we have not encountered a single instance of an orbital subperiosteal abscess in a patient with orbital cellulitis in either the outpatient or inpatient setting. However, we have encountered isolated cases of soft tissue abscess in the area of the lacrimal sac or in connection with large flaps used for reconstruction; these have generally been localized to a heavily surgically treated and irradiated area near the nasal cavity or other part of the paranasal sinuses. Thus, the absence of a subperiosteal abscess in a patient with suspected orbital cellulitis and a history of malignancy should not put the clinician’s mind at ease.
Treatment of Orbital Cellulitis in Cancer Patients
As soon as orbital cellulitis is suspected in a cancer patient because of clinical findings of periorbital soft tissue edema or erythema, empirical broad-spectrum antibacterial and antifungal treatments should be initiated. The use of broad-spectrum bactericidal or fungicidal agents given intravenously in appropriate doses is of paramount importance. This therapy should be instituted before imaging studies of the orbit are arranged and before ophthalmology and ear, nose, and throat consultations are obtained. In the majority of patients, broad-spectrum antibiotics are started empirically; subsequently, careful clinical evaluation indicates whether de-escalation or further intensification of antibiotics is needed. Finally, it is important to correct underlying metabolic abnormalities in patients with orbital cellulitis (e.g., to obtain glucose control in patients with uncontrolled hyperglycemia). Figures 13.3 and 13.4 depict a reasonable approach to the empirical choice of anti-infectives in patients with cancer and suspected orbital cellulitis at our institution. This approach, however, is not necessarily applicable to every patient.
Empiric treatment of acute orbital cellulitis in patients with solid tumors
Empiric treatment of acute orbital cellulitis in patients with leukemia and/or transplant recipients*
In view of the relative rarity of orbital cellulitis in cancer patients, there is no consensus regarding the optimal management of this condition. Treatment is individualized and depends on factors such as the presence of comorbidities (especially kidney or renal dysfunction), drug-drug interactions, prior exposure to antibiotics, and the possibility of selection for resistant organisms and local epidemiology of resistance. In addition, the choice of antibiotic is individualized and depends on the rapidity of symptom onset (rapid onset favors bacterial etiology), the presence of features suggestive of adjacent fungal sinusitis, and the possibility for intraorbital or even central nervous system (e.g., cavernous sinus thrombosis) involvement.
CT or MRI of orbit to evaluate the extent of cellulitis and to check for the presence of a drainable focus is critical. Blood culture should be performed in patients with septicemia and secondary “seeding” to the orbit, and the use of biomarkers (beta-D glucan, Aspergillus galactomannan) might be helpful in selected patients. Biopsy is reserved for rare cases in which tumor infiltration masquerading as infection (discussed in the next section of the chapter) cannot be ruled out.
An infectious disease consultation is highly recommended in patients with severe or atypical orbital cellulitis or in patients in whom resistant strains of bacteria are suspected.
Regarding the need for surgical intervention, cancer patients with orbital cellulitis often have accompanying thrombocytopenia due to cancer treatments such as chemotherapy or bone marrow transplant. Thus, surgical intervention for orbital or sinus infection is often not an early consideration as it might be in healthier patients. Obviously, should an abscess actually be present, it should be managed acutely with surgical drainage. However, as we have already established, a subperiosteal abscess is a very rare finding in cancer patients with orbital cellulitis. The current philosophy is to employ watchful watching after rapid initiation of broad-spectrum antibiotics and reserve surgery for selected patients who have an intraorbital abscess or necrotic lesions in the sinuses.
Malignancies can present with signs and symptoms suggestive of orbital cellulitis. For example, there are numerous reports of cancer masquerading as invasive fungal sinusitis with signs of cavernous sinus involvement, a scenario seemingly overrepresented in patients with T-cell lymphomas [27–33]. Lesions such as Langerhans cell histiocytosis have also been treated as orbital cellulitis before the correct diagnosis was made . Additionally, the scientific literature contains multiple reports of intraocular or orbital tumors and metastases in the orbit or paranasal sinuses that presented the like and were initially misdiagnosed as orbital cellulitis [35–42]. Retinoblastoma can masquerade as orbital cellulitis, as indicated in multiple reports [43–46]; so can pleomorphic adenoma . Interestingly, there are also reports of orbital masses initially believed to be tumors but later confirmed to be chronic infectious abscesses .
More recently, noninfectious orbital inflammatory conditions have emerged that are related to treatment for cancer. Subtenon injection of carboplatin for the treatment of retinoblastoma has been associated with noninfectious orbital inflammation . Bisphosphonates, often used in cancer patients with abnormal bony metabolism, have long been known to cause various types of ocular or orbital inflammation [50, 51]. Intravesical bacille Calmette-Guérin immunotherapy used for bladder cancer has been reported to cause orbital inflammation . Such scenarios are expected to become more frequent as oncologists treat more diseases with immune-modulating therapies. It is therefore prudent in all cases of suspected orbital cellulitis in patients with a history of cancer to note of all treatments rendered for their disease.