Evaluation of Subconjunctival Triamcinolone for Nonnecrotizing Anterior Scleritis

Evaluation of Subconjunctival Triamcinolone for Nonnecrotizing Anterior Scleritis Thomas A. Albini, MD,1 Ehud Zamir, MD,2 Russell W. Read, MD,3 Ronald E. Smith, MD,1 Robert F. See, MD,1 Narsing A. Rao, MD1 Purpose: Subconjunctival corticosteroid injection (SCI) for nonnecrotizing anterior scleritis remains controversial, partly because long-term follow-up is not available. This study documents the efficacy and adverse effects of SCI. Design: Retrospective, noncomparative, interventional case series. Participants: Thirty-eight eyes of 35 patients with nonnecrotizing, noninfectious anterior scleritis resistant to prior systemic or local therapy. Intervention: Subconjunctival triamcinolone acetonide injection. Main Outcome Measure: Persistence or resolution of signs and symptoms of anterior scleritis and development of complications. Results: Thirty-six of 38 eyes had complete resolution of signs and symptoms within 6 weeks of SCI. Fifteen eyes had follow-up of ⱖ30 months. There were no instances of scleral melting or perforation; adverse events included subconjunctival hemorrhage (5 patients), transient ocular hypertension without evidence of glaucoma (4 patients), cataract (2 patients), and glaucoma (2 patients). Subconjunctival corticosteroid injection resulted in reduced dependence on systemic medications. Conclusion: Subconjunctival corticosteroid injection in eyes that failed other therapies is effective, reduces dependence on systemic medications, and did not result in scleral necrosis over a median follow-up period of 29 months. Ophthalmology 2005;112:1814 –1820 © 2005 by the American Academy of Ophthalmology.

Anterior scleritis is a sight-threatening ocular inflammation, characterized by acute or chronic red eye and ocular pain caused by inflammation directed against the sclera. Anterior scleritis may be due to infectious agents, it may be related to systemic autoimmune diseases, or it may be idiopathic. When associated with a systemic autoimmune disease, the most common found is rheumatoid arthritis. Treatment of noninfectious anterior scleritis has traditionally included the use of systemically administered nonsteroidal antiinflammatory drugs (NSAIDs), corticosteroids, immunosuppresOriginally received: December 20, 2004. Accepted: May 9, 2005. Manuscript no. 2004-424. 1 Doheny Eye Institute, Keck School of Medicine, University of Southern California, Los Angeles, California. 2 Department of Ophthalmology, Hadassah Medical Organization, Jerusalem, Israel. 3 Department of Ophthalmology, University of Alabama School of Medicine, University of Alabama at Birmingham, Birmingham, Alabama. Data presented at: American Academy of Ophthalmology Annual Meeting, October, 2004; New Orleans, Louisiana. Supported by an unrestricted grant from Research to Prevent Blindness, New York, New York. The authors have no proprietary interest in any of the materials used in the study. Reprint requests and correspondence to Narsing A. Rao, Doheny Eye Institute, 1450 San Pablo Street, DVRC 211, Los Angeles, CA 90033. E-mail: [email protected].

1814

© 2005 by the American Academy of Ophthalmology Published by Elsevier Inc.

sive agents, or a combination.1,2 Although these systemic medications may achieve adequate control of scleritis, they carry with them the possibility for significant morbidity due to the adverse effects of treatment. The present study expands on a previous prospective series of subconjunctival corticosteroid injection (SCI) for nonnecrotizing anterior scleritis in 12 eyes of 10 patients,3 providing longer patient follow-up of these original patients and data from 25 additional patients treated with SCI.

Patients and Methods Institutional review board approval was obtained for this study. This study is a retrospective, noncomparative, interventional case series. Eligibility criteria for SCI included the persistence of active, nonnecrotizing, noninfectious anterior scleritis despite systemic or local antiinflammatory therapy. Risks and benefits of SCI and systemic therapies were carefully explained to patients, with emphasis on previous reports of scleral necrosis attributed to SCI, as well as the risks of subconjunctival hemorrhage, ocular hypertension, cataract, glaucoma, and other possible untoward effects. All participating patients gave informed consent for SCI. Before April 2002, only patients who had experienced failure of systemic antiinflammatory medicines were recruited. Because SCI was effective in a previous prospective study without scleral necrosis,3 beginning in April 2002 we recruited additional patients who had failed local immunomodulatory therapy (topical corticosteroids or SCI administered elsewhere). We excluded patients with ocular hypertension (defined as an intraocular pressure ISSN 0161-6420/05/$–see front matter doi:10.1016/j.ophtha.2005.05.008

Albini et al 䡠 Subconjunctival Triamcinolone for Nonnecrotizing Anterior Scleritis [IOP] of ⬎22 mmHg without ocular antihypertensive therapy), a history of glaucoma (determined by progressive visual field [VF] changes and/or optic disc changes consistent with glaucomatous optic neuropathy), or a history of steroid response (defined as IOP increased to ⬎22 mmHg in either eye, after local steroid therapy). Patients with necrotizing scleritis or necrotizing keratitis were also excluded. All patients were referred to 1 of 4 tertiary care centers: Doheny Eye Institute; the uveitis service at Los Angeles County/University of Southern California Hospital; Department of Ophthalmology, University of Alabama at Birmingham; or the Hadassah Medical Organization. All patients underwent a detailed evaluation, including ophthalmic and systemic history, slit-lamp examination, dilated fundus examination, and medical and laboratory evaluations for underlying systemic abnormalities. Scleritis was defined as a painful red eye with congestion of the deep episcleral venous plexus in conjunction with scleral edema, both evident after blanching of the conjunctival vessels by 10% phenylephrine. Necrotizing features were defined as thinning and destruction of the sclera, evidence of vascular closure on red-free examination, and/or frank uveal prolapse. If inflammatory control was achieved after injection, then systemic medications were tapered gradually according to best medical judgment. In cases of an underlying rheumatologic disease, systemic medication adjustments were made in coordination with the treating rheumatologist or internist based on the needs of the systemic condition. Resolution was defined as lack of any pain, episcleral injection, or scleral edema. The data recorded for each patient included age, race, gender, the involved eye, associated autoimmune disorder, therapy used before SCI, therapies used after SCI, resolution of signs and symptoms by the 6-week visit after SCI, patients’ subjective report of time to resolution after SCI, duration of follow-up, recurrence, and adverse effects. Therapies both before and after SCI were further analyzed by tabulating how many patients used each of 3 forms of therapy (NSAIDs, prednisone, immunosuppressive drugs) with or without other medications. Kaplan–Meier survival estimates were used to analyze the recurrence of anterior scleritis in eyes treated with SCI. The technique of SCI was detailed in our previous report.3 Briefly, the affected eyes were anesthetized with topical 0.5% proparicaine, applied with drops and a cotton-tipped applicator placed on the area of injection for 30 seconds. A 25-gauge needle was used to inject triamcinolone acetonide (Kenalog-40, BristolMyers Squibb, Princeton, NJ) under the conjunctiva directly over the inflamed areas. In eyes with nodular scleritis, 0.05 to 0.1 ml was injected immediately adjacent to the nodule; areas of diffuse scleritis received 0.2 ml per inflamed quadrant. All patients had full ophthalmic examinations 2 weeks and 6 weeks after injection and at 1- to 3-month intervals thereafter. Patients who developed ocular hypertension were further evaluated by a glaucoma specialist. Steroid-response glaucoma was defined as an IOP of ⬎22 mmHg found in conjunction with VF abnormalities and/or optic disc changes.

and prior SCI performed by the referring ophthalmologist in the other. Thirty-six of 38 eyes (94%) experienced complete resolution of symptoms after a single treatment by the 6-week follow-up (Table 1). Two eyes experienced only partial resolution after SCI. Azathioprine was added in one case and ibuprofen in the other, leading to complete resolution of scleritis in both cases and subsequent discontinuation of systemic treatment. There were no cases of scleral necrosis or detectable scleral thinning after SCI (Fig 1). Adverse events included subconjunctival hemorrhages (5 eyes); transient ocular hypertension without evidence of glaucoma (4 eyes); cataract (2 eyes); and glaucoma (2 eyes), requiring medical management in one case and surgical management in the other. In all cases of ocular hypertension or glaucoma, patient management was continued in conjunction with a glaucoma specialist. In one patient with ocular hypertension, surgical excision of the subconjunctival triamcinolone depot was performed, with a return to normal IOP. The overall median follow-up was 29 months (range, 1– 48). Nine eyes developed a recurrence of scleritis during the follow-up. The median time to recurrence was 11 months. Of these 9 eyes, 4 were from patients with idiopathic disease, and 5 were from patients with associated autoimmune disease. Of the 15 eyes with ⬎30 months’ follow-up, 6 experienced recurrence at a median of 22.5 months. Using Kaplan–Meier life-table analysis of 38 eyes treated with SCI, recurrence of scleritis occurred in 11.7% at 6 months, 18.2% at 13 months, 26.4% at 32 months, and 55.8% at 41 months (Fig 2). Subconjunctival corticosteroid injection reduced the need for systemic therapy, particularly in patients with idiopathic scleritis. Before SCI, 33 of 35 patients (94%) were treated with systemic antiinflammatory medication; 21 patients used NSAIDs, 21 used prednisone, and 9 used immunosuppressive drugs (typically patients used more than one of these therapies). After SCI, 17 of 35 patients (49%) continued to require systemic antiinflammatory medication; 4 used NSAIDs, 9 used prednisone, and 7 used immunosuppressive drugs. Systemic medication was used after SCI because of contralateral scleritis in a patient with a history of steroid-response glaucoma after SCI (1/17), only partial response to SCI (2/17), recurrent symptoms of scleritis (5/17), or treatment of associated autoimmune disease (9/17). Five patients required systemic therapy for recurrent symptoms after SCI: 1 in whom ocular hypertension prohibited further SCI; 2 with early recurrence of scleritis (at 1 and 4 months) who were subsequently successfully treated with systemic therapy; 1 with idiopathic scleritis treated first with repeat SCI for a recurrence at 41 months, without complete resolution, necessitating the addition of systemic prednisone; and 1 with idiopathic scleritis who experienced complete resolution with SCI but later developed mild periocular pain without frank scleritis, responding to a cyclooxygenase-2 inhibitor. Of 19 patients with idiopathic scleritis in the study, 6 (32%) used systemic medications after SCI (2 because of only partial response to SCI, 1 because of glaucoma, 1 because of incomplete resolution of scleritis after repeat SCI, 1 for early recurrence, and 1 for mild periocular pain without frank scleritis).

Results A total of 38 eyes of 35 patients with noninfectious nonnecrotizing scleritis were treated with SCI between August 1999 and July 2004 (Table 1). The mean age of all patients was 46 years at the time of injection. The male-to-female ratio was 1:3. Of the 35 patients, 15 were Caucasian; 14, Hispanic; 3, African or African American; 2, Asian; and 1, Asian Indian. Nineteen patients had idiopathic disease. Sixteen patients had an associated autoimmune condition. Thirty-six eyes had failed systemic therapy before SCI. Two eyes had failed prior local therapy: topical corticosteroids in one case

Discussion The most commonly used treatment strategy for nonnecrotizing noninfectious anterior scleritis has been thoroughly reviewed by Jabs et al.2 The complications of a standardized, stepped, systemic therapy for scleritis are documented by the authors in a retrospective chart review including data on standard treatment and response to standard treatment of

1815

Ophthalmology Volume 112, Number 10, October 2005 Table 1. Patient Age (yrs)

Race

Gender

Institution

Associated Autoimmune Disorder

Eye

Scleritis DA DA DA

Seronegative arthritis

40

Hispanic

M

DEI

43

Asian

F

DEI

Right Left Left

42

Hispanic

F

DEI

Left

NA

Idiopathic

40

Hispanic

F

DEI

Right

DA

Rheumatoid arthritis

40

Hispanic

F

DEI

Left

NA

Rheumatoid arthritis

44

Asian Indian

F

DEI

Right

NA

Idiopathic

Left

NA

Idiopathic

24

Hispanic

F

DEI

Right

NA

Idiopathic

50

Caucasian

M

DEI

Left

DA

Idiopathic

59 62

Hispanic Caucasian

F M

DEI DEI

Left Right

NA DA

Rheumatoid arthritis Psoriasis and relapsing polychondritis

64

Hispanic

M

DEI

Left Left

DA NA

MCTD and rosacea

35

Caucasian

M

DEI

Left

DA

Relapsing polychondritis

57 55 70

Hispanic Asian Hispanic

F F F

DEI DEI DEI

Right Right Left

DA DA DA

Idiopathic Idiopathic Rheumatoid arthritis

30

Caucasian

M

DEI

Right

DA

Wegener granulomatosis

43 63 47 57

Hispanic Hispanic Hispanic Hispanic

F F F F

LAC LAC LAC LAC

Right Right Left Left

DA DA NA NA

Idiopathic Idiopathic Idiopathic Relapsing polychondritis

32

Caucasian

F

HMO

Right

DA

SLE

18 72 55 52 45 38

Caucasian Caucasian Caucasian Caucasian African Caucasian

F M F F F F

HMO HMO HMO HMO HMO HMO

Left Right Left Left Right Left

DA DA NA DA NA DA

Idiopathic Idiopathic Rheumatoid arthritis, PUK Rosacea SLE Idiopathic

34 42 48

Caucasian African American Caucasian

F M F

HMO UAB UAB

Right Right Right

NA NA DA

Erythema nodosum, IBD Idiopathic IBD

49

African American

M

UAB

Right

DA

Idiopathic

24

Hispanic

F

UAB

Right

DA

Idiopathic

48 32 53

Caucasian Caucasian Caucasian

F F F

UAB UAB UAB

Right Right Right

DA DA DA

Idiopathic Idiopathic Idiopathic

Cox-2 ⫽ cyclooxygenase 2; CSA ⫽ cyclosporine A; DA ⫽ diffuse anterior; DEI ⫽ Doheny Eye Institute; HMO ⫽ Hadassah Medical Organization; IBD ⫽ methotrexate; NA ⫽ nodular anterior; OHT ⫽ ocular hypertension; PUK ⫽ peripheral ulcerative keratitis; SCI ⫽ subconjunctival corticosteroid injection;

1816

Albini et al 䡠 Subconjunctival Triamcinolone for Nonnecrotizing Anterior Scleritis Characteristics Prior Therapy

Therapy after SCI

Indomethacin, prednisone

Prednisone, MTX

Indomethacin, prednisone, azathioprine Indomethacin, prednisone

None Azathioprine

Naproxan, prednisone, CSA, None MTX, etanercept, hydroxychloroquine, celecoxib Indomethacin, celecoxib, None prednisone Ibuprofen, prednisone, Prednisone azathioprine

Resolution of Symptoms and Signs Complete Complete Complete Partial, added azathioprine Complete

Months to First Recurrence 39

Retreatment SCI

11

SCI

Indomethacin, prednisone Prednisone, azathioprine Indomethacin Indomethacin Prednisone, azathioprine, infliximab Cyclophosphamide Indomethacin Indomethacin Prednisone Prednisone

42

41

SCI, prednisone

45

Complete

None

Complete

35

None CSA

Complete Complete Complete

34 34 33

Complete

31

Complete

31

Complete Complete Complete

28 28 2

Complete

13

Complete Complete Complete Complete

10 11 18 17

Prednisone, cyclophosphamide, infliximab None None Prednisone Ibuprofen

Prednisone, cyclophosphamide Indomethacin Indomethacin Indomethacin Celecoxib Prednisone Topical corticosteroids

Hydroxychloroquine Complete None None Indomethacin None Prednisone Cox-2 inhibitor

Complete Complete Complete Complete Complete Complete

Ibuprofen Ibuprofen Prednisone

None None None

Complete Complete Complete

Prednisone

Complete

Ibuprofen

MTX for contralateral scleritis Ibuprofen

SCI elsewhere Prednisone, MTX Prednisone

None None Prednisone, MTX

Partial, added ibuprofen Complete Complete Complete

Cataract

42

None

Prednisone, azathioprine Prednisone, infliximab None None Prednisone

OHT OHT

30

Complete Indomethacin, prednisone, azathioprine Cox-2 inhibitors, prednisone, orbital floor steroid injection Prednisone, azathioprine Ibuprofen, prednisone

41 41 48

Adverse Effects

3

Complete Complete

Months of Follow-up

32

1

6

4

13

Topical corticosteroids

Prednisone

Hydroxychloroquine

SCI

SCI

38

34 30 29 29 30 23 21 24 22 34 4 17

4

Prednisone, MTX

Subconjunctival hemorrhage

Subconjunctival hemorrhage

OHT

Glaucoma requiring maximal medical management OHT, excision of depot

Subconjunctival hemorrhage Subconjunctival hemorrhage Cataract, glaucoma requiring surgery Subconjunctival hemorrhage

1 12 7

inflammatory bowel disease; LAC ⫽ Los Angeles County/University of Southern California Hospital; MCTD ⫽ mixed connective tissue disease; MTX ⫽ SLE ⫽ systemic lupus erythematosus; UAB ⫽ University of Alabama at Birmingham.

1817

Ophthalmology Volume 112, Number 10, October 2005

Figure 1. External photographs of the left eye of a 63-year-old male with relapsing polychondritis and diffuse nonnecrotizing anterior scleritis previously treated with subconjunctival corticosteroid injection (SCI) in the fellow eye. Photographs were taken before SCI (A, D), 3 weeks after SCI (B, E), and 33 months after SCI (C, F). Arrows indicate subconjunctival steroid deposits. Photographs taken 33 months after SCI do not show any evidence of scleral thinning or necrosis in the areas of prior SCI.

69 patients with scleritis. The first line of treatment was oral NSAIDs. Approximately 67% of patients did not respond to NSAIDs alone. Patients with nonnecrotizing anterior scleritis who did not respond to NSAIDs were started on oral prednisone at 1 mg/kg/day. This dose was continued until the scleritis was quiet for 1 month. The prednisone was then slowly tapered, yielding a median of 30 weeks’ duration of prednisone therapy for patients who did respond to prednisone. In 26.1% of cases, immunosuppressive agents were used for a median of 1 year along with prednisone for a median of 44 weeks. The complications of prednisone included hyperglycemia (10%), corticosteroid psychosis (7%), and osteoporosis (5%). The complications of immunosuppressive therapy included leukopenia (defined by the authors as a leukocyte count below 2500 cells/␮l) (37%),

microscopic hematuria (19%), gross hematuria (6.2%), cardiomyopathy (6%), cytomegalovirus pneumonitis (6%), and Pneumocytis carinii pneumonia. Orbital floor steroids have been used as an adjunctive therapy to reduce scleral inflammation transiently4; however, the need for frequent repeat injection has limited their use. Subconjunctival corticosteroid injection has been used with longer-lasting effects for nonnecrotizing anterior scleritis but remains controversial.5,6 Some authors consider this therapy to be ineffective and contraindicated because of a perceived risk of inducing scleral necrosis.1,7 The literature concerning this therapeutic approach was summarized in our recent prospective evaluation of 12 eyes undergoing SCI.3 The direct evidence that is cited to suggest that SCI results in necrotizing complications consists of 6 cases

Figure 2. Kaplan–Meier estimate of time to recurrence of anterior scleritis in 38 eyes treated with subconjunctival corticosteroid injection.

1818

Albini et al 䡠 Subconjunctival Triamcinolone for Nonnecrotizing Anterior Scleritis reported in non–peer reviewed literature.6 –9 These brief reports provide limited documentation of the type of corticosteroids used, location of injection, or type of scleritis. In these reports, a cause and effect relationship between the treatment and the complication has not been established. In contrast, 2 retrospective series describing a total of 29 patients10,11 and our previously published prospective series describing 10 patients3 found an encouraging therapeutic response without necrotizing complications. One admitted limitation of our original prospective study was the relatively small number of patients with a limited follow-up period.6 We have attempted to address this limitation via the current series, in which 38 eyes of 35 patients with nonnecrotizing anterior scleritis underwent SCI. This resulted in complete resolution of the signs and symptoms of scleritis within 6 weeks in 36 of 38 eyes (94%). The efficacy of SCI in this series is similar to that reported in other retrospective series.10,11 Although this noncomparative study cannot be used to compare the efficacy of SCI to that of standard therapy directly, the efficacy of SCI seems to compare favorably to that reported for both systemic therapy and orbital floor steroids.3 In the 2 cases in the current study that achieved only a partial response to SCI, complete resolution was achieved with the subsequent addition of systemic medications. It is important to point out that all patients except one in this series used other treatments before SCI. Although prior immunosuppression was not successful in controlling scleritis in any of these eyes, its beneficial effects cannot be excluded when considering the efficacy of SCI in this series. Subconjunctival corticosteroid injection resulted in marked decreased dependence on systemic medications. Fifty percent of patients treated with SCI required no systemic medication after SCI. In most instances when systemic medication was required after SCI, it was for the treatment of associated autoimmune conditions. Sixty-eight percent of patients with idiopathic scleritis treated with SCI did not require systemic medication thereafter. Therefore, SCI provided excellent efficacy and spared patients the added risk of complications resulting from months of systemic therapy. Patients in this series were observed for a median of 29 months, including 15 patients who were observed for ⬎30 months. We identified complications of SCI, including subconjunctival hemorrhage (13.1%), transient ocular hypertension (10.5%), cataract (5.2%), and glaucoma (5.2%), reinforcing that it is essential to carefully exclude patients with glaucoma or a history of steroid response. Glaucoma was the only irreversible complication of SCI. Because patients with ocular hypertension and glaucoma in this study were treated largely by different glaucoma specialists, there was no standardized approach to treatment of this complication when IOP could not be controlled by medical management. It has been reported that as many as 13% of scleritis patients treated by standard therapy develop glaucoma, mostly due to inflammatory damage to the trabecular meshwork.12 Because many of these patients are using oral and/or topical corticosteroids, standard treatment of scleritis also exposes patients to a risk of steroid-response glaucoma. Although the proximity of the triamcinolone depot to the

trabecular meshwork suggests that steroid-response glaucoma could be more prevalent with SCI than with systemic or topical therapy, this noncomparative study cannot be used to draw conclusions about the relative risk of steroidinduced glaucoma with SCI as compared with standard therapies for scleritis. No patients developed scleral necrosis in this series or in the 2 previous retrospective series in the literature.10,11 Although definitive proof of scleral melting consequent to SCI does not exist, it remains difficult to prove clinically that scleral necrosis is an acceptably rare complication of SCI. Applying the rule of three13 (i.e., if none of x patients develop a complication, then we can be reasonably confident—95%— that the true rate of this event is no more than 3/x) suggests that the rate of scleral melting after SCI is no more than 8%. This figure can be lowered to 5% if one includes the 29 patients reported in prior series.10,11 The question remains whether this upper limit of risk of a potentially devastating complication without definitive proof of its occurrence should preclude the use of SCI and thereby expose patients to the known morbidity, length of therapy, and well-documented complications of standard therapy. Informed patient consent with the physican’s careful explanation and the patient’s demonstration of understanding of the relative risks involved is essential. This study was designed to gather increasing evidence of the efficacy and safety of SCI for nonnecrotizing anterior scleritis in eyes that failed other therpaies. It is not comparative and provides no direct evidence of the relative efficacy or complications of SCI as compared with standard therapy. It documents complications of SCI, including glaucoma. Careful selection of patients for SCI and careful patient follow-up are required. Until a randomized comparison is available, this study provides further evidence that SCI is rapidly effective and reduces the number of systemic medications used and, consequently, the risk of complications of systemic steroids and/or immunosuppressive agents. Acknowledgments. The authors thank Laurie Labree, statistician and project manager at the Department of Preventive Medicine, Keck School of Medicine, University of Southern California, for Kaplan–Meier analysis.

References 1. Sainz de la Maza M, Jabbur NS, Foster CS. An analysis of therapeutic decision for scleritis. Ophthalmology 1993;100: 1372– 6. 2. Jabs DA, Mudun A, Dunn JP, Marsh MJ. Episcleritis and scleritis: clinical features and treatment results. Am J Ophthalmol 2000;130:469 –76. 3. Zamir E, Read RW, Smith RE, et al. A prospective evaluation of subconjunctival injection of triamcinolone acetonide for resistant anterior scleritis. Ophthalmology 2002; 109:798 – 805. 4. Hakin KN, Ham J, Lightman SL. Use of orbital floor steroids in the management of patients with uniocular non-necrotising scleritis. Br J Ophthalmol 1991;75:337–9. 5. Cunningham ET Jr. Discussion [of Zamir E, Read RW, Smith RE, et al. A prospective evaluation of subconjunctival injection of triamcinolone acetonide for resistant anterior scleritis]. Ophthalmology 2002;109:805– 6.

1819

Ophthalmology Volume 112, Number 10, October 2005 6. Jabs DA. Discussion [of Zamir E, Read RW, Smith RE, et al. A prospective evaluation of subconjunctival injection of triamcinolone acetonide for resistant anterior scleritis]. Ophthalmology 2002;109:806 –7. 7. Watson PG, Hazleman BL. The Sclera and Systemic Disorders. Philadelphia: Saunders, 1976:409. 8. Watson PG, Hazleman BL, Pavésio CE, Green WR. The Sclera and Systemic Disorders. 2nd ed. New York: ButterworthHeinemann, 2004:295. 9. Watson PG. Treatment of scleritis and episcleritis. Trans Ophthalmol Soc U K 1974;94:76 –9.

1820

10. Tu EY, Culbertson WW, Pflugfelder SC, et al. Therapy of nonnecrotizing anterior scleritis with subconjunctival corticosteroid injection. Ophthalmology 1995;102:718 – 24. 11. Croasdale CR, Brightbill FS. Subconjunctival corticosteroid injections for nonnecrotizing anterior scleritis. Arch Ophthalmol 1999;117:966 – 8. 12. Sainz de la Maza M, Jabbur NS, Foster CS. Severity of scleritis and episcleritis. Ophthalmology 1994;101:389 –96. 13. Schachat AP, Chambers WA, Liesegang TJ, Albert DA. Safe and effective. Ophthalmology 2003;110:2073– 4.