Blindness resulting from orbital cornpIications of-si nusitis BRADFORD S. PATT, MD. and SCOTT C. MANNING, MD, Dallas, Texas
Loss of vision remains a potential complication of orbital infection. Appropriate evaluation and management of the patient with signs and symptoms of orbital inflammation may prevent progression to blindness. Evaluation of patients with orbital inflammation from sinusitis includes a comprehensive clinical examination and radiographic studies. Clinical examination should test for changes in visual acuity, pupillary reactivity, and extraocular motion. Computerized tomography (CT) has facilitated the diagnosis of orbital infections and aids in diagnosis. However, CT can be misleading in patients with acute orbital infections and should not be relied on to determine the need for surgical intervention. We reviewed the records of all patients admitted to Parkland Memorial Hospital from 1978 to 1988 with orbital complications resulting from sinusitis. Four of 159 patients in this group had permanent blindness. The presence of an abscess, which was ultimately found at surgical exploration, was not diagnosed by CT in any of these four patients. Clinical examination remains the most important indicator for surgical intervention in patients with orbital complications of sinusitis. We present our findings and give guidelines for surgical intervention in patients with orbital infections resulting from Sinusitis. (OTOLARYNGOL HEAD NECK SURG 1991;104:789.)
T h e clinical terms periorbital and orbital c e h l i t i s encompass a wide range of anatomic and etiologic possibilities and can therefore be confusing. Orbital infections are more clearly defined in relation to the orbital septum, which is the anterior extension of the periosteum of the orbit. Inflammation anterior to the orbital septum or preseptal cellulitis is common in young children. It is often associated with Hemophilus injuenzae septicemia, although it may result from penetrating injury or from secondary infection of skin lesions. It rarely involves postseptal anatomy, and physical examination thus reveals lid edema in the absence of orbital signs such as gaze restriction and propto~is.~ Treatment consists of appropriate broad-spectrum antibiotics, and resolution is usually rapid. Surgery is not indicated except in the case of progression to lid ab-
From the Department of Otorhinolaryngology, University of Texas Southwestern Medical Center. Presented at the Annual Meeting of the American Academy of Otolaryngology-Head and Neck Surgery. New Orleans, La., Sept. 24-28, 1989. Received for publication Nov. 3, 1989; revision received Nov. 16, 1990; accepted Nov. 21, 1990. Reprint requests: Scott C . Manning, MD, Department of Otorhinolaryngology, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235-9035. 231 I I26897
scess. Inflammation behind the orbital septum or postsepta1 cellulitis can be divided anatomically by the extraocular muscles into extraconol and intraconal disease.4 Local spread of inflammation from adjacent infected sinuses is by far the most common cause of postseptal infection. The most common entity in this category is subperiosteal inflammatory disease adjacent to infected ethmoid sinuses.2 Infection may spread via dehiscence in the lamina papyracea or via septic thrombophlebiti~.~ Intraconal postseptal cellulitis can result from further progression of extraconal disease or from penetrating trauma or seeding of septic emboli.* Postseptal cellulitis in general is seen with signs of chemosis, proptosis, and restricted gaze. Clinical examination alone cannot distinguish early inflammation from abscess or extraconal from intraconal disease.6As the severity of signs and symptoms increases, however, so does the likelihood of abscess formation with the need for surgical drainage. Progression to ophthalmoplegia and loss of vision specifically suggests the likelihood of intraconal abscess, although subperiosteal extraconal abscesses can duplicate these findings via pressure effect. Cavernous sinus thrombosis represents the most severe form of postseptal cellulitis and results from extension of disease via valveless veins between the orbit and the cavernous sinus. Cavernous sinus thrombosis 789
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
OtolaryngologyHead and Neck Surgery
790 PAT and MANNING
220 PATIENTS- ACUTE SINUSITIS ADMISSIONS
159 Orbital Complications
70 - Postseptal Inflammation
89 - Preseptal Inflammation
30 - Subperiosteal
Abscess
2 - Orbital Abscess
1 - Blindness
3 - Blindness Cases
Case
Fig. 1. Breakdown of Parkland Hospital admissions, 1978 to 1988.
Table 1. Characteristics of blindness patients Patient no.
Age (yrllsex
1
44/F
Predisposing factors
Diabetes mellitus
Visual acuity at presentation
OD 20/20
0s 20/60 2
77/F
Diabetes mellitus Upper respiratory infection
OD 20/20 0s LPO
15/M
Swimming and diving
Not tested
25/M
Molar carie Upper respiratory infection
OD 20/20 0s 20/50
CT findings
Surgical findings
Left medial subOpacification of sinuses "with no eviperiosteal abdence of orbital inscess fection" Edema of extraocular Left orbital abmuscles with inflamscess rnation, "no abscess" within orbit Orbital proptosis OS, Left subperiosteal "no intraorbital abscess mass" Edema of retroorbital Left subperiosteal abscess tissues, "no mass effect to suggest abscess"
Visual acuity at followup
0 s : LPO
0s:LPO 0s:NLP 0s: LPO
P O , Light perception only, NLP. no light perception
is suggested clinically by bilateral disease with ophthalmoplegia and loss of vision. Permanent loss of vision resulting from orbital inflammation is unusual in this era of antibiotics, but it still O C C U ~ Timely S . ~ surgical drainage of postseptal abscesses and of adjacent infected sinuses may prevent the sequelae of blindness. Because clinical examination cannot by itself exactly delineate the nature of postsepta1 inflammatory processes, clinicians have increas-
ingly come to rely on computed tomography (CT) to select appropriate patients for surgery. When CT scan demonstrates strong evidence of postseptal abscess formation, few investigators would argue against surgical drainage.8 The problem arises when radiographs demonstrate areas of postseptal mass effect, contrast enhancement, or both without definite tissue liquefaction. In these patients, the clinician must rely on the clinical progression of the inflammation via visual acuity test-
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Volume 104 Number 6 June 1992
Blindness resulting from orbital complications of sinusitis 791
ing, pupillary reactivity, and gaze restriction assessment.9.’0 A review of the records from Parkland Memorial Hospital over a 10-year period has yielded four cases of permanent unilateral blindness as a result of postseptal orbital inflammation. In each case, CT readings of “no definite abscess” contributed to a delay in diagnosis of orbital suppuration, with a resultant delay in surgical drainage.
METHODS AND MATERIALS From 1978 to 1988, a total of 256 patients were admitted to Parkland Memorial Hospital with the diagnosis of acute sinusitis. O f these 256 charts, 220 were available for review. Orbital complications occurred in 159 of the 220 patients, including 89 patients with presepta1 cellulitis, 38 patients with postseptal cellulitis, 30 patients with subperiosteal abscess, and two patients with orbital abscess as a final diagnosis (Fig. 1). All cases of preseptal cellulitis resolved with intravenous antibiotic therapy. Among the 70 patients with postseptal cellulitis, subperiosteal abscess, or orbital abscess, 60 underwent CT scanning of the sinuses and orbits with a GE 8800 o r Picker scanner (Picker Company, Grand Prairie, Texas). All CT scans were reviewed by a staff radiologist. Surgical drainage of the sinuses and exploration of the affected orbit were performed in 32 of the 70 patients, at which time the definitive diagnosis of postseptal cellulitis, subperiosteal abscess, or ortibal abscess was made. At surgical exploration, 10 of 60 CT scans were found to have been misinterpreted from their initial reading of absence of orbital abscess or subperiosteal abscess. Irreversible unilateral blindness occurred in 4 of these 10 patients over this 10-year period (Table 1). CASE REPORTS
Case 1. A 44-year-old woman with diabetes was admitted to the Parkland Hospital ophthalmology service with a 1-day history of pain and swelling of the left eye. The patient also had had purulent nasal discharge for 7 days before admission. On initial examination, the patient had left orbital proptosis, chemosis, and restricted mobility in upward and downward gaze. Temperature on admission was 38.2” C, and the white blood cell count was 17,700. Results of an ophthalmologic examination revealed a visual acuity of 20/20 in the right eye and 20160 in the left eye. CT axial scan of the sinuses and orbits showed opacification of both ethmoid and maxillary sinuses bilaterally, with no definite evidence of an orbital or subperiosteal abscess (Fig. 2). The patient was started on a regimen of intravenous cefuroxime and nasal saline irrigations. Eighteen hours later, the patient continued to have proptosis, and the visual acuity was noted to have deteriorated to 201 100 in the affected eye, with a corresponding afferent pupillary defect. Surgical exploration of the orbit was performed through an external ethmoidectomy approach, and a well-formed subperiosteal abscess was drained. Postopera-
tively, the proptosis and gaze restriction resolved, but visual acuity has remained at light perception only. Case 2. A 77-year-old woman with diabetes was first seen at the Parkland Memorial Hospital Ophthalmology service with a 1-week history of upper respiratory infection and a new onset of swelling and tenderness of the left eye. On examination, she was diaphoretic, febrile, and hypertensive. Results of ophthalmologic examination revealed left orbital proptosis and ophthalmoplegia with intact pupillary reflexes. Visual acuity was reduced to light perception only. CT of the sinuses and orbits showed edema of the extraocular muscles, but no definite abscess formation within the bony walls of the orbit. The left ethmoid, frontal, sphenoid, and maxillary sinuses were opacified without gas fluid level. The patient was treated initially with intravenous methicillin and chloramphenicol. No improvement was noted after 48 hours of intravenous antibiotic therapy. Repeat CT scan of the orbits and sinuses showed no abscess formation. After 72 hours, the patient had continued proptosis with no improvement in vision and a sluggish pupillary reflex. Surgical exploration was performed, which revealed an orbital abscess and a necrotic medial rectus muscle. The patient has light perception only in the left eye after 6 years of followup. Case 3. A 15-year-old previously healthy boy had acute onset of malaise, fever, and chills associated with swelling in the left eye. He had been swimming and diving 2 days earlier. On examination at a local hospital, his left eye was proptotic and “massively” swollen. No visual acuity testing was performed because of “swelling of the eye.” Plain sinus films showed left maxillary opacification. The patient was transferred to a university medical center where examination of the eye revealed proptosis, chemosis, severe gaze restriction in all fields, and “diminished” left pupillary reflex and light perception visual acuity. CT scan showed orbital proptosis, but no definite abscess within the orbit (Fig. 3). There was left-sided ethmoid and maxillary sinus opacification. Ophthalmologic examination revealed a proptotic, ophthalmoplegic left eye with no light perception. After 48 hours of intravenous antibiotic therapy with no improvement in visual acuity, a definite afferent pupillary defect was noted. The patient then underwent a left external ethmoidectomy with drainage of a left subperiosteal abscess. He has not regained vision after 2 years of followup. Case 4. A otherwise healthy 25-year-old man was admitted to Parkland Memorial Hospital with a 2-day history of left-sided facial and orbital pain with associated edema. The patient had had a foul nasal discharge over the previous 2 weeks with a painful left molar, but no changes in vision. On initial evaluation, the patient was febrile with a white blood cell count of 30,500. Ophthalmologic examination revealed left orbital proptosis and chemosis. Gaze was mildly restricted in all directions, and the eye was deviated laterally and inferiorly. Visual acuity was 20120 in the right eye and 20150 in the left eye initially. Computerized tomography of the sinuses and orbits showed left ethmoid and maxillary sinus opacification. The retroorbital tissues were edematous, but no definite abscess was seen (Fig. 4), and the patient was started on a regimen of
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
OtolaryngologyHead a n d N e c k Surgery
792 PAll and MANNING
Fig. 2. Case 1.A 44-year-oldwoman with edema of extraocular muscles, but no evidence of abscess.
Fig. 3 . Case 3. A 15-year-oldboy with extensive proptosis and lateralization of medial rectus muscle, but equivocal for abscess formation.
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Volume 104 Number 6
Blindness resulting from orbital complications of sinusitis 793
June 1991
Fig. 4. Case 4. A 25-year-oldman with retro-orbitaledema and proptosis, but no definite abscess formation.
intravenous penicillin. tobramycin, and chloramphenicol, The carious left maxillary molar was extractcd. and the patient underwent aspiration of thc left maxillary sinus with subscquent growth of Stciplrylococcirs ~ i ~ r c w sOver . the next 24 hours, serial visual examinations showed rapid deterioration of vision in the left eyc to 20/200. with further gaze restriction and a new finding of a hypereniic optic disk. Pupillary reflexes remained normal. Orbital exploration through a left external ethmoidectomy approach revealed a subpcriosteal abscess. This was drained along with the cthmoid and left maxillary sinus via a Caldwcll-Luc approach. Forty-eight hours latcr the patient required repeat orbital exploration because of continued orbital proptosis with no improvement in vision. No further purulence was located within the orbit at that time. The patient had gradual resolution of the facial and orbital edema, but has had no improvement in vision after 4 years of followup.
DISCUSSION
Permanent loss of vision has been noted as a complication of orbital infection since 1893," and up to 20% of patients with postseptal inflamniation had blindness in the pre-antibiotic era." In the present series, 4 of 38 patients with postseptal disease incurred permanent loss of vision. This 10.5% incidence is in line with other reports in the recent literature.' '' Obviously, broad-spectrum antibiotics and modern surgical tech-
niques have not eliminated the potential for this devastating problem. The mechanism for loss of vision with orbital inflammation may involve ( 1 ) optic neuritis as a reaction to adjacent or nearby infection, (2) ischemia resulting from thrombophlebitis along valveless orbital veins, or (3) pressure ischemia possibly resulting in central retinal artery occlusion. Pressure ischemia is especially relevant to subperiosteal abscesses, which can arise suddenly and rapidly increase intraocular pressure to dangerous levels. Relief of intraocular pressure elevations is no doubt the cause of restoration of vision after surgical drainage in reported cases of subperiosteal abscess with preoperative loss of vision.' Unfortunately. as the present case series demonstrates, sometimes retinal damage becomes irreversible before definitive treatment of the orbital inflammation occurs. I t is the job of the clinician to use every means available to decide if and when surgery is indicated, to try to prevent or reverse loss of vision and other potential sequelae of orbital cellulitis. Computed tomography has been an invaluable addition in the diagnostic evaluation of orbital disease. It has facilitated the diagnosis of orbital inflammation by allowing excellent resolution of the globe. retro-orbital tissues, adjacent sinuses. and intracranial contents. An
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
OtolaryngologyHead and Neck Surgery
794 PAT and MANNING
abscess is suggested on CT scan in general by the presence of a low-density mass effect, with or without rim enhancement.I6A gas-fluid level within a mass is very specific, but this finding is often absent with abscesses later confirmed at surgery. Lateral displacement of the medial rectus muscle with associated edema is an indication of subperiosteal inflammation, but subperiosteal abscess can only be identified with certainty in the face of very low density or gas adjacent to the lamina papyracea. CT scan is unfortunately limited in the degree to which it can delineate differences in soft tissue densities, especially in acute inflammatory s t a t e ~As .~ this series demonstrates, the finding of “no definite abscess’’ can be misleading if not put in the perspective of CT limitations. Most investigators agree that CT is unnecessary for evaluation of preseptal periorbital cellulitis; that is, when there is no evidence of proptosis or gaze restriction.16 They also agree that CT is indicated when neurologic signs are present, to rule out associated epidural or subdural empyema, brain abscess, or cavernous sinus thrombosis.l5 In addition, most investigators recommend CT when signs of postseptal inflammation are progressing for 24 to 48 hours despite treatment with broad-spectrum antibiotics, but the value of CT in directing treatment is controversial.6 Hirsch and Lifshitz” have advocated the use of CT as the primary factor in determining therapy in cases of orbital inflammation. Gutowski et al.,” on the other hand, found that CT tended to underestimate the incidence of orbital suppuration in a series of 33 patients with ethmoiditis and postseptal inflammation.l8 Although there were no false-positive diagnoses of abscess, they noted a few false-negative findings because of the inability of CT to always resolve abscess from cellulitis. Although CT findings were in agreement with the final diagnosis in 82% of patients in the series, they had little effect on the outcome because decisions with regard to therapy generally were directed by clinical progression of orbital signs. The physical signs of postseptal orbital inflammation include proptosis, gaze restriction, diminished pupillary reflexes, and decreased visual acuity. Certain patterns of signs are strongly suggestive of specific diagnoses, such as the association between lateral proptosis with decreased adduction and medial subperiosteal abscess. But orbital signs in general are nonspecific, although increase in severity is associated with increasing likelihood of a post-septa1suppurativeprocess. Visual acuity and funduscopic and pupillary findings can be associated with increased intraocular pressure in general, but are not specific enough to allow for a definitive diagnosis.2
In the face of nonspecific physical signs and potentially false-negative CT findings, most investigators have stressed use of the progression of clinical findings to determine when surgical intervention is necessary. If orbital signs do not improve, or worsen after 24 to 48 hours of appropriate antibiotic therapy, surgery is recommended with CT as a guide to orbital and sinus drainage. Problems arise in a clinical setting, however, when (1) absence of definite abscess on CT scan is used to justify prolonging medical treatment in the face of deteriorating orbital signs, (2) serial clinical ophthalmologic examination is difficult because of lack of patient cooperation, massive periorbital edema, or both, or (3) overemphasis is placed on easier to identify “objective”findings, such as diminished pupillary reflexes, which may not be seen until irreversible loss of vision has occurred. In the present series, 4 of 38 patients with postseptal orbital cellulitis had permanent loss of vision. In each patient, negative or equivocal CT findings contributed to a delay in surgical intervention, although it is impossible to know for certain whether early or surgical drainage of the sinuses and orbits would have resulted in a return of vision. The first two patients had diabetes and were initially seen with significantly decreased visual acuity. In retrospect, because of their immunosuppression, a case could be made for draining the involved sinuses and exploring the adjacent orbital walls at the outset in these patients. Case 3 had no light perception, an ophthalmologic emergency, with associated pansinusitis. Again, more appropriate management would have been immediate drainage of the sinus with exploration of the adjacent orbit. The fourth patient was allowed to demonstrate steady deterioration in visual acuity over a 24-hour period before the ear, nose, and throat service was consulted. Again, in retrospect, sinus and orbital drainage should have occurred at the very first sign of orbital progression. All decisions for surgical intervention involved the risk-benefit analysis of the “whole patient” and management paradigms are invariably too simplistic to apply to every person. For example, in the case of postsepta1 orbital inflammation, some investigators recommend treating subperiosteal abscess nonsurgically if orbital signs resolve steadily with antibiotic therapy. l9 However, as the present series demonstrates, the risk of blindness with orbital suppuration is still very real in this era of antibiotic therapy, and surgical drainage is still the quickest and most definitive method of addressing the problem when the diagnosis is confirmed by CT or strongly suggested by clinical and radiographic findings. Clinicians must pay strict attention to the initial clinical presentation, progression of orbital signs, and overall health of the patient, without being
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Volume 104 Number 6 June 1991
Blindness resulting from orbital complications of sinusitis 795
led astray by lack of definite suppuration on CT.’ On the basis of the Parkland Hospital experience of blindness resulting from orbital complications of sinusitis, we recommend surgical exploration of involved sinuses and adjacent orbital anatomy in cases of (1) definite CT evidence of postseptal orbital suppuration, (2) initial presentation of significantly decreased visual acuity (20/60) with sinusitis in an immunocompromised patient, (3) initial presentation of severe orbital compromise, such as blindness or afferent pupillary defect with ipsilateral sinusitis, and (4) any significant progression of orbital signs while on a regimen of intravenous antibiotics. REFERENCES 1. Jackson K, Baker S. Clinical implications of orbital cellulitis.
Laryngoscope 1986;96:568-74. 2. Harris G. Subperiostal abscess of the orbit. Arch Ophthalmol 1983;101:751-7. 3. Weiss A, Friendly D, Eglin K, Chang M, Gold B. Bacterial periorbital and orbital cellulitis in childhood. Ophthalmology 1983;90:195-203. 4. Towbin R, Han B, Kaufmann R, Burke M. Postseptal cellulitis: CT in diagnosis and management. Radiology 1986;158:735-7. 5. Jarrett W, Gutman FA. Ocular complications of infection in the paranasal sinuses. Arch Ophthalmol 1969;81:683-8. 6. Schramm V, Curtin H, Kennerdell JS. Evaluation of orbital cellulitis and results of treatment. Laryngoscope 1982;92:732-8. 7. Schramm VL, Myers EN, Kennerdell JS. Orbital complications of acute sinusitis evaluation, management, and outcome. Otolaryngology 1978;86:221-30.
8. Goodwin WJ, Weinshall M, Chandler J. The role of high resolution computerized tomography and standardized ultrasound in the evaluation of orbital cellulitis. Laryngoscope 1982;92:72831. 9. Moloney JR, Badham NJ, McRae A. The acute orbit: preseptal (periorbital) cellulitis, subperiostal abscess, and orbital cellulitis due to sinusitis. J Laryngol Otol 1987;12:1-18. 10. Gold SC, Arvigg PG, Hedges TR. Computerized tomography in the management of acute orbital cellulitis. Ophthalmic Surg 1987;18:753-6. 11. Jackson K, Baker SR. Periorobital cellulitis. Head Neck Surg 1987;2:227-34. 12. Trantas A. Opbtbalmoplegie totale, et autres complications oculaires, dans les polysinusities. Arch Ophthalmol 1893;13:357. 13. Gamble RC. Acute inflammation of the orbit in children. Arch Ophthalmol 1933;10:453-97. 14. Duke-Elder S, MacFaul PA. System of ophthalmology. St. Louis: CV Mosby Company, 1974:859-89. 15. Zimmermann RA, Bilaniuki CT.CT of orbital infection and its cerebral complications. Am J Roentgengol 1980;134:45-9. 16. Weber AL, Mikuli D. Inflammatory disorders of the periorbital sinuses and their complications. Radiol Clin North Am 1987; 25:615-30. 17. Hirsch M, Lifshitz T. Computerized tomography in diagnosis and treatment of orbital cellulitis. Pediatr Radiol 1988;18: 302-5. 18. Gutowski WM, Mulbury E, Hengerer AS, Kid0 DK. The role of CT scans in managing the orbital complications of ethmoiditis. Int J Pediatr Otorhinolaryngol 1988;15:117-28. 19. Souliere C, Gregory A, Michael M, Blumberg A, Isaacson G. Selective non-surgical management of subperiostal abscess of the orbit: computerized tomography and clinical course as indication for surgical drainage. Int J Pediatr Otorhinolaryngol 1990;19: 109-19.
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Loss of vision remains a potential complication of orbital infection. Appropriate evaluation and management of the patient with signs and symptoms of orbital inflammation may prevent progression to blindness. Evaluation of patients with orbital inflammation from sinusitis includes a comprehensive clinical examination and radiographic studies. Clinical examination should test for changes in visual acuity, pupillary reactivity, and extraocular motion. Computerized tomography (CT) has facilitated the diagnosis of orbital infections and aids in diagnosis. However, CT can be misleading in patients with acute orbital infections and should not be relied on to determine the need for surgical intervention. We reviewed the records of all patients admitted to Parkland Memorial Hospital from 1978 to 1988 with orbital complications resulting from sinusitis. Four of 159 patients in this group had permanent blindness. The presence of an abscess, which was ultimately found at surgical exploration, was not diagnosed by CT in any of these four patients. Clinical examination remains the most important indicator for surgical intervention in patients with orbital complications of sinusitis. We present our findings and give guidelines for surgical intervention in patients with orbital infections resulting from Sinusitis. (OTOLARYNGOL HEAD NECK SURG 1991;104:789.)
T h e clinical terms periorbital and orbital c e h l i t i s encompass a wide range of anatomic and etiologic possibilities and can therefore be confusing. Orbital infections are more clearly defined in relation to the orbital septum, which is the anterior extension of the periosteum of the orbit. Inflammation anterior to the orbital septum or preseptal cellulitis is common in young children. It is often associated with Hemophilus injuenzae septicemia, although it may result from penetrating injury or from secondary infection of skin lesions. It rarely involves postseptal anatomy, and physical examination thus reveals lid edema in the absence of orbital signs such as gaze restriction and propto~is.~ Treatment consists of appropriate broad-spectrum antibiotics, and resolution is usually rapid. Surgery is not indicated except in the case of progression to lid ab-
From the Department of Otorhinolaryngology, University of Texas Southwestern Medical Center. Presented at the Annual Meeting of the American Academy of Otolaryngology-Head and Neck Surgery. New Orleans, La., Sept. 24-28, 1989. Received for publication Nov. 3, 1989; revision received Nov. 16, 1990; accepted Nov. 21, 1990. Reprint requests: Scott C . Manning, MD, Department of Otorhinolaryngology, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd., Dallas, TX 75235-9035. 231 I I26897
scess. Inflammation behind the orbital septum or postsepta1 cellulitis can be divided anatomically by the extraocular muscles into extraconol and intraconal disease.4 Local spread of inflammation from adjacent infected sinuses is by far the most common cause of postseptal infection. The most common entity in this category is subperiosteal inflammatory disease adjacent to infected ethmoid sinuses.2 Infection may spread via dehiscence in the lamina papyracea or via septic thrombophlebiti~.~ Intraconal postseptal cellulitis can result from further progression of extraconal disease or from penetrating trauma or seeding of septic emboli.* Postseptal cellulitis in general is seen with signs of chemosis, proptosis, and restricted gaze. Clinical examination alone cannot distinguish early inflammation from abscess or extraconal from intraconal disease.6As the severity of signs and symptoms increases, however, so does the likelihood of abscess formation with the need for surgical drainage. Progression to ophthalmoplegia and loss of vision specifically suggests the likelihood of intraconal abscess, although subperiosteal extraconal abscesses can duplicate these findings via pressure effect. Cavernous sinus thrombosis represents the most severe form of postseptal cellulitis and results from extension of disease via valveless veins between the orbit and the cavernous sinus. Cavernous sinus thrombosis 789
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
OtolaryngologyHead and Neck Surgery
790 PAT and MANNING
220 PATIENTS- ACUTE SINUSITIS ADMISSIONS
159 Orbital Complications
70 - Postseptal Inflammation
89 - Preseptal Inflammation
30 - Subperiosteal
Abscess
2 - Orbital Abscess
1 - Blindness
3 - Blindness Cases
Case
Fig. 1. Breakdown of Parkland Hospital admissions, 1978 to 1988.
Table 1. Characteristics of blindness patients Patient no.
Age (yrllsex
1
44/F
Predisposing factors
Diabetes mellitus
Visual acuity at presentation
OD 20/20
0s 20/60 2
77/F
Diabetes mellitus Upper respiratory infection
OD 20/20 0s LPO
15/M
Swimming and diving
Not tested
25/M
Molar carie Upper respiratory infection
OD 20/20 0s 20/50
CT findings
Surgical findings
Left medial subOpacification of sinuses "with no eviperiosteal abdence of orbital inscess fection" Edema of extraocular Left orbital abmuscles with inflamscess rnation, "no abscess" within orbit Orbital proptosis OS, Left subperiosteal "no intraorbital abscess mass" Edema of retroorbital Left subperiosteal abscess tissues, "no mass effect to suggest abscess"
Visual acuity at followup
0 s : LPO
0s:LPO 0s:NLP 0s: LPO
P O , Light perception only, NLP. no light perception
is suggested clinically by bilateral disease with ophthalmoplegia and loss of vision. Permanent loss of vision resulting from orbital inflammation is unusual in this era of antibiotics, but it still O C C U ~ Timely S . ~ surgical drainage of postseptal abscesses and of adjacent infected sinuses may prevent the sequelae of blindness. Because clinical examination cannot by itself exactly delineate the nature of postsepta1 inflammatory processes, clinicians have increas-
ingly come to rely on computed tomography (CT) to select appropriate patients for surgery. When CT scan demonstrates strong evidence of postseptal abscess formation, few investigators would argue against surgical drainage.8 The problem arises when radiographs demonstrate areas of postseptal mass effect, contrast enhancement, or both without definite tissue liquefaction. In these patients, the clinician must rely on the clinical progression of the inflammation via visual acuity test-
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Volume 104 Number 6 June 1992
Blindness resulting from orbital complications of sinusitis 791
ing, pupillary reactivity, and gaze restriction assessment.9.’0 A review of the records from Parkland Memorial Hospital over a 10-year period has yielded four cases of permanent unilateral blindness as a result of postseptal orbital inflammation. In each case, CT readings of “no definite abscess” contributed to a delay in diagnosis of orbital suppuration, with a resultant delay in surgical drainage.
METHODS AND MATERIALS From 1978 to 1988, a total of 256 patients were admitted to Parkland Memorial Hospital with the diagnosis of acute sinusitis. O f these 256 charts, 220 were available for review. Orbital complications occurred in 159 of the 220 patients, including 89 patients with presepta1 cellulitis, 38 patients with postseptal cellulitis, 30 patients with subperiosteal abscess, and two patients with orbital abscess as a final diagnosis (Fig. 1). All cases of preseptal cellulitis resolved with intravenous antibiotic therapy. Among the 70 patients with postseptal cellulitis, subperiosteal abscess, or orbital abscess, 60 underwent CT scanning of the sinuses and orbits with a GE 8800 o r Picker scanner (Picker Company, Grand Prairie, Texas). All CT scans were reviewed by a staff radiologist. Surgical drainage of the sinuses and exploration of the affected orbit were performed in 32 of the 70 patients, at which time the definitive diagnosis of postseptal cellulitis, subperiosteal abscess, or ortibal abscess was made. At surgical exploration, 10 of 60 CT scans were found to have been misinterpreted from their initial reading of absence of orbital abscess or subperiosteal abscess. Irreversible unilateral blindness occurred in 4 of these 10 patients over this 10-year period (Table 1). CASE REPORTS
Case 1. A 44-year-old woman with diabetes was admitted to the Parkland Hospital ophthalmology service with a 1-day history of pain and swelling of the left eye. The patient also had had purulent nasal discharge for 7 days before admission. On initial examination, the patient had left orbital proptosis, chemosis, and restricted mobility in upward and downward gaze. Temperature on admission was 38.2” C, and the white blood cell count was 17,700. Results of an ophthalmologic examination revealed a visual acuity of 20/20 in the right eye and 20160 in the left eye. CT axial scan of the sinuses and orbits showed opacification of both ethmoid and maxillary sinuses bilaterally, with no definite evidence of an orbital or subperiosteal abscess (Fig. 2). The patient was started on a regimen of intravenous cefuroxime and nasal saline irrigations. Eighteen hours later, the patient continued to have proptosis, and the visual acuity was noted to have deteriorated to 201 100 in the affected eye, with a corresponding afferent pupillary defect. Surgical exploration of the orbit was performed through an external ethmoidectomy approach, and a well-formed subperiosteal abscess was drained. Postopera-
tively, the proptosis and gaze restriction resolved, but visual acuity has remained at light perception only. Case 2. A 77-year-old woman with diabetes was first seen at the Parkland Memorial Hospital Ophthalmology service with a 1-week history of upper respiratory infection and a new onset of swelling and tenderness of the left eye. On examination, she was diaphoretic, febrile, and hypertensive. Results of ophthalmologic examination revealed left orbital proptosis and ophthalmoplegia with intact pupillary reflexes. Visual acuity was reduced to light perception only. CT of the sinuses and orbits showed edema of the extraocular muscles, but no definite abscess formation within the bony walls of the orbit. The left ethmoid, frontal, sphenoid, and maxillary sinuses were opacified without gas fluid level. The patient was treated initially with intravenous methicillin and chloramphenicol. No improvement was noted after 48 hours of intravenous antibiotic therapy. Repeat CT scan of the orbits and sinuses showed no abscess formation. After 72 hours, the patient had continued proptosis with no improvement in vision and a sluggish pupillary reflex. Surgical exploration was performed, which revealed an orbital abscess and a necrotic medial rectus muscle. The patient has light perception only in the left eye after 6 years of followup. Case 3. A 15-year-old previously healthy boy had acute onset of malaise, fever, and chills associated with swelling in the left eye. He had been swimming and diving 2 days earlier. On examination at a local hospital, his left eye was proptotic and “massively” swollen. No visual acuity testing was performed because of “swelling of the eye.” Plain sinus films showed left maxillary opacification. The patient was transferred to a university medical center where examination of the eye revealed proptosis, chemosis, severe gaze restriction in all fields, and “diminished” left pupillary reflex and light perception visual acuity. CT scan showed orbital proptosis, but no definite abscess within the orbit (Fig. 3). There was left-sided ethmoid and maxillary sinus opacification. Ophthalmologic examination revealed a proptotic, ophthalmoplegic left eye with no light perception. After 48 hours of intravenous antibiotic therapy with no improvement in visual acuity, a definite afferent pupillary defect was noted. The patient then underwent a left external ethmoidectomy with drainage of a left subperiosteal abscess. He has not regained vision after 2 years of followup. Case 4. A otherwise healthy 25-year-old man was admitted to Parkland Memorial Hospital with a 2-day history of left-sided facial and orbital pain with associated edema. The patient had had a foul nasal discharge over the previous 2 weeks with a painful left molar, but no changes in vision. On initial evaluation, the patient was febrile with a white blood cell count of 30,500. Ophthalmologic examination revealed left orbital proptosis and chemosis. Gaze was mildly restricted in all directions, and the eye was deviated laterally and inferiorly. Visual acuity was 20120 in the right eye and 20150 in the left eye initially. Computerized tomography of the sinuses and orbits showed left ethmoid and maxillary sinus opacification. The retroorbital tissues were edematous, but no definite abscess was seen (Fig. 4), and the patient was started on a regimen of
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
OtolaryngologyHead a n d N e c k Surgery
792 PAll and MANNING
Fig. 2. Case 1.A 44-year-oldwoman with edema of extraocular muscles, but no evidence of abscess.
Fig. 3 . Case 3. A 15-year-oldboy with extensive proptosis and lateralization of medial rectus muscle, but equivocal for abscess formation.
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Volume 104 Number 6
Blindness resulting from orbital complications of sinusitis 793
June 1991
Fig. 4. Case 4. A 25-year-oldman with retro-orbitaledema and proptosis, but no definite abscess formation.
intravenous penicillin. tobramycin, and chloramphenicol, The carious left maxillary molar was extractcd. and the patient underwent aspiration of thc left maxillary sinus with subscquent growth of Stciplrylococcirs ~ i ~ r c w sOver . the next 24 hours, serial visual examinations showed rapid deterioration of vision in the left eyc to 20/200. with further gaze restriction and a new finding of a hypereniic optic disk. Pupillary reflexes remained normal. Orbital exploration through a left external ethmoidectomy approach revealed a subpcriosteal abscess. This was drained along with the cthmoid and left maxillary sinus via a Caldwcll-Luc approach. Forty-eight hours latcr the patient required repeat orbital exploration because of continued orbital proptosis with no improvement in vision. No further purulence was located within the orbit at that time. The patient had gradual resolution of the facial and orbital edema, but has had no improvement in vision after 4 years of followup.
DISCUSSION
Permanent loss of vision has been noted as a complication of orbital infection since 1893," and up to 20% of patients with postseptal inflamniation had blindness in the pre-antibiotic era." In the present series, 4 of 38 patients with postseptal disease incurred permanent loss of vision. This 10.5% incidence is in line with other reports in the recent literature.' '' Obviously, broad-spectrum antibiotics and modern surgical tech-
niques have not eliminated the potential for this devastating problem. The mechanism for loss of vision with orbital inflammation may involve ( 1 ) optic neuritis as a reaction to adjacent or nearby infection, (2) ischemia resulting from thrombophlebitis along valveless orbital veins, or (3) pressure ischemia possibly resulting in central retinal artery occlusion. Pressure ischemia is especially relevant to subperiosteal abscesses, which can arise suddenly and rapidly increase intraocular pressure to dangerous levels. Relief of intraocular pressure elevations is no doubt the cause of restoration of vision after surgical drainage in reported cases of subperiosteal abscess with preoperative loss of vision.' Unfortunately. as the present case series demonstrates, sometimes retinal damage becomes irreversible before definitive treatment of the orbital inflammation occurs. I t is the job of the clinician to use every means available to decide if and when surgery is indicated, to try to prevent or reverse loss of vision and other potential sequelae of orbital cellulitis. Computed tomography has been an invaluable addition in the diagnostic evaluation of orbital disease. It has facilitated the diagnosis of orbital inflammation by allowing excellent resolution of the globe. retro-orbital tissues, adjacent sinuses. and intracranial contents. An
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
OtolaryngologyHead and Neck Surgery
794 PAT and MANNING
abscess is suggested on CT scan in general by the presence of a low-density mass effect, with or without rim enhancement.I6A gas-fluid level within a mass is very specific, but this finding is often absent with abscesses later confirmed at surgery. Lateral displacement of the medial rectus muscle with associated edema is an indication of subperiosteal inflammation, but subperiosteal abscess can only be identified with certainty in the face of very low density or gas adjacent to the lamina papyracea. CT scan is unfortunately limited in the degree to which it can delineate differences in soft tissue densities, especially in acute inflammatory s t a t e ~As .~ this series demonstrates, the finding of “no definite abscess’’ can be misleading if not put in the perspective of CT limitations. Most investigators agree that CT is unnecessary for evaluation of preseptal periorbital cellulitis; that is, when there is no evidence of proptosis or gaze restriction.16 They also agree that CT is indicated when neurologic signs are present, to rule out associated epidural or subdural empyema, brain abscess, or cavernous sinus thrombosis.l5 In addition, most investigators recommend CT when signs of postseptal inflammation are progressing for 24 to 48 hours despite treatment with broad-spectrum antibiotics, but the value of CT in directing treatment is controversial.6 Hirsch and Lifshitz” have advocated the use of CT as the primary factor in determining therapy in cases of orbital inflammation. Gutowski et al.,” on the other hand, found that CT tended to underestimate the incidence of orbital suppuration in a series of 33 patients with ethmoiditis and postseptal inflammation.l8 Although there were no false-positive diagnoses of abscess, they noted a few false-negative findings because of the inability of CT to always resolve abscess from cellulitis. Although CT findings were in agreement with the final diagnosis in 82% of patients in the series, they had little effect on the outcome because decisions with regard to therapy generally were directed by clinical progression of orbital signs. The physical signs of postseptal orbital inflammation include proptosis, gaze restriction, diminished pupillary reflexes, and decreased visual acuity. Certain patterns of signs are strongly suggestive of specific diagnoses, such as the association between lateral proptosis with decreased adduction and medial subperiosteal abscess. But orbital signs in general are nonspecific, although increase in severity is associated with increasing likelihood of a post-septa1suppurativeprocess. Visual acuity and funduscopic and pupillary findings can be associated with increased intraocular pressure in general, but are not specific enough to allow for a definitive diagnosis.2
In the face of nonspecific physical signs and potentially false-negative CT findings, most investigators have stressed use of the progression of clinical findings to determine when surgical intervention is necessary. If orbital signs do not improve, or worsen after 24 to 48 hours of appropriate antibiotic therapy, surgery is recommended with CT as a guide to orbital and sinus drainage. Problems arise in a clinical setting, however, when (1) absence of definite abscess on CT scan is used to justify prolonging medical treatment in the face of deteriorating orbital signs, (2) serial clinical ophthalmologic examination is difficult because of lack of patient cooperation, massive periorbital edema, or both, or (3) overemphasis is placed on easier to identify “objective”findings, such as diminished pupillary reflexes, which may not be seen until irreversible loss of vision has occurred. In the present series, 4 of 38 patients with postseptal orbital cellulitis had permanent loss of vision. In each patient, negative or equivocal CT findings contributed to a delay in surgical intervention, although it is impossible to know for certain whether early or surgical drainage of the sinuses and orbits would have resulted in a return of vision. The first two patients had diabetes and were initially seen with significantly decreased visual acuity. In retrospect, because of their immunosuppression, a case could be made for draining the involved sinuses and exploring the adjacent orbital walls at the outset in these patients. Case 3 had no light perception, an ophthalmologic emergency, with associated pansinusitis. Again, more appropriate management would have been immediate drainage of the sinus with exploration of the adjacent orbit. The fourth patient was allowed to demonstrate steady deterioration in visual acuity over a 24-hour period before the ear, nose, and throat service was consulted. Again, in retrospect, sinus and orbital drainage should have occurred at the very first sign of orbital progression. All decisions for surgical intervention involved the risk-benefit analysis of the “whole patient” and management paradigms are invariably too simplistic to apply to every person. For example, in the case of postsepta1 orbital inflammation, some investigators recommend treating subperiosteal abscess nonsurgically if orbital signs resolve steadily with antibiotic therapy. l9 However, as the present series demonstrates, the risk of blindness with orbital suppuration is still very real in this era of antibiotic therapy, and surgical drainage is still the quickest and most definitive method of addressing the problem when the diagnosis is confirmed by CT or strongly suggested by clinical and radiographic findings. Clinicians must pay strict attention to the initial clinical presentation, progression of orbital signs, and overall health of the patient, without being
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
Volume 104 Number 6 June 1991
Blindness resulting from orbital complications of sinusitis 795
led astray by lack of definite suppuration on CT.’ On the basis of the Parkland Hospital experience of blindness resulting from orbital complications of sinusitis, we recommend surgical exploration of involved sinuses and adjacent orbital anatomy in cases of (1) definite CT evidence of postseptal orbital suppuration, (2) initial presentation of significantly decreased visual acuity (20/60) with sinusitis in an immunocompromised patient, (3) initial presentation of severe orbital compromise, such as blindness or afferent pupillary defect with ipsilateral sinusitis, and (4) any significant progression of orbital signs while on a regimen of intravenous antibiotics. REFERENCES 1. Jackson K, Baker S. Clinical implications of orbital cellulitis.
Laryngoscope 1986;96:568-74. 2. Harris G. Subperiostal abscess of the orbit. Arch Ophthalmol 1983;101:751-7. 3. Weiss A, Friendly D, Eglin K, Chang M, Gold B. Bacterial periorbital and orbital cellulitis in childhood. Ophthalmology 1983;90:195-203. 4. Towbin R, Han B, Kaufmann R, Burke M. Postseptal cellulitis: CT in diagnosis and management. Radiology 1986;158:735-7. 5. Jarrett W, Gutman FA. Ocular complications of infection in the paranasal sinuses. Arch Ophthalmol 1969;81:683-8. 6. Schramm V, Curtin H, Kennerdell JS. Evaluation of orbital cellulitis and results of treatment. Laryngoscope 1982;92:732-8. 7. Schramm VL, Myers EN, Kennerdell JS. Orbital complications of acute sinusitis evaluation, management, and outcome. Otolaryngology 1978;86:221-30.
8. Goodwin WJ, Weinshall M, Chandler J. The role of high resolution computerized tomography and standardized ultrasound in the evaluation of orbital cellulitis. Laryngoscope 1982;92:72831. 9. Moloney JR, Badham NJ, McRae A. The acute orbit: preseptal (periorbital) cellulitis, subperiostal abscess, and orbital cellulitis due to sinusitis. J Laryngol Otol 1987;12:1-18. 10. Gold SC, Arvigg PG, Hedges TR. Computerized tomography in the management of acute orbital cellulitis. Ophthalmic Surg 1987;18:753-6. 11. Jackson K, Baker SR. Periorobital cellulitis. Head Neck Surg 1987;2:227-34. 12. Trantas A. Opbtbalmoplegie totale, et autres complications oculaires, dans les polysinusities. Arch Ophthalmol 1893;13:357. 13. Gamble RC. Acute inflammation of the orbit in children. Arch Ophthalmol 1933;10:453-97. 14. Duke-Elder S, MacFaul PA. System of ophthalmology. St. Louis: CV Mosby Company, 1974:859-89. 15. Zimmermann RA, Bilaniuki CT.CT of orbital infection and its cerebral complications. Am J Roentgengol 1980;134:45-9. 16. Weber AL, Mikuli D. Inflammatory disorders of the periorbital sinuses and their complications. Radiol Clin North Am 1987; 25:615-30. 17. Hirsch M, Lifshitz T. Computerized tomography in diagnosis and treatment of orbital cellulitis. Pediatr Radiol 1988;18: 302-5. 18. Gutowski WM, Mulbury E, Hengerer AS, Kid0 DK. The role of CT scans in managing the orbital complications of ethmoiditis. Int J Pediatr Otorhinolaryngol 1988;15:117-28. 19. Souliere C, Gregory A, Michael M, Blumberg A, Isaacson G. Selective non-surgical management of subperiostal abscess of the orbit: computerized tomography and clinical course as indication for surgical drainage. Int J Pediatr Otorhinolaryngol 1990;19: 109-19.
Downloaded from oto.sagepub.com at CAMBRIDGE UNIV LIBRARY on October 28, 2015
