Classification and management of Coats disease: the 2000 Proctor Lecture

Classification and Management of Coats Disease: The 2000 Proctor Lecture JERRY A. SHIELDS, MD, CAROL L. SHIELDS, MD, SANTOSH G. HONAVAR, MD, HAKAN DEMIRCI, MD, AND JACQUELINE CATER, PHD

● PURPOSE:

To review the methods and results of management in a large series of patients with Coats disease, to determine risk factors for poor visual outcome and enucleation, and to propose a practical classification of Coats disease. ● METHODS: In a retrospective consecutive series in 150 patients, Coats disease was defined as idiopathic retinal telangiectasia with intraretinal or subretinal exudation without appreciable signs of retinal or vitreal traction. We reviewed our experience with management, including observation, laser photocoagulation, cryotherapy, and various techniques of retinal detachment surgery and See also pp. 561–571.

enucleation. The anatomic outcome, complications of treatment, visual results, and reasons for enucleation were tabulated. Factors predictive of poor visual outcome (20/200 or worse) and enucleation were determined using Cox proportional hazards regression models. Based on these observations, a staging classification of Coats disease, applicable to treatment selection and ocular prognosis, is proposed. ● RESULTS: In 117 patients (124 eyes) with a mean follow up of 55 months (range, 6 months to 25 years) primary management was observation in 22 eyes (18%), cryotherapy in 52 (42%), laser photocoagulation in 16 (13%), various methods of retinal detachment surgery in 20 (17%), and enucleation in 14 (11%). Anatomic Accepted for publication Jan 18, 2001. From the Oncology Service, Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania. This study was presented by Dr Jerry Shields as part of the 2000 Proctor Lecture, University of California San Francisco School of Medicine, San Francisco, California, December 7, 2000. This work was supported by the Eye Tumor Research Foundation, Philadelphia, Pennsylvania, the Award of Merit in Retina Research, Houston, Texas (Dr J. Shields), the Macula Foundation, New York, New York (Dr C. Shields), the Hyderabad Eye Research Foundation, Hyderabad, India (Dr Honavar), and Orbis International, New York, New York (Dr Honavar). Reprint requests to Jerry A. Shields, MD, Oncology Service, Wills Eye Hospital, 900 Walnut St, Philadelphia, PA 19107.

572

©

2001 BY

improvement or stability was achieved in 76% of eyes, and final visual acuity was 20/50 or better in 17 eyes (14%), 20/60 to 20/100 in eight (6%), 20/200 to finger counting in 30 (24%), and hand motion to no light perception in 49 (40%) Enucleation was ultimately necessary in 20 eyes (16%). Risk factors predictive of poor visual outcome (20/200 or worse) included postequatorial (P ⴝ .01), diffuse (P ⴝ .01), or superior (P ⴝ .04) location of the telangiectasias and exudation, failed resolution of subretinal fluid after treatment (P ⴝ .02), and presence of retinal macrocysts (P ⴝ .02). The main risk factors for enucleation were elevated intraocular pressure (greater than 22 mm Hg; P less than or equal to .001) and iris neovascularization (P less than or equal to .001). Coats disease was classified into stage 1, telangiectasia only; stage 2, telangiectasia and exudation (2A, extrafoveal exudation; 2B, foveal exudation) stage 3, exudative retinal detachment (3A, subtotal; 3B, total); stage 4, total detachment and secondary glaucoma; and stage 5, advanced end-stage disease. Poor visual outcome (20/200 or worse) was found in 0% of eyes with stage 1, 53% with stage 2, 74% with stage 3, and 100% of stages 4 and 5 Coats disease. Enucleation was ultimately necessary in 0% of stages 1 and 2, 7% of stage 3, 78% of stage 4, and 0% of stage 5 disease. ● CONCLUSIONS: Carefully selected treatment can anatomically stabilize or improve the eye with Coats disease in 76% of eyes. However, poor visual outcome of 20/200 or worse commonly results. Patients who present with stages 1 to 3 Coats disease have the best visual prognosis, and patients with stages 4 and 5 have a poor visual prognosis. (Am J Ophthalmol 2001;131:572–583. © 2001 by Elsevier Science Inc. All rights reserved.)

C

OATS DISEASE IS A NONHEREDITARY CONDITION

that is characterized by idiopathic retinal telangiectasia, exudation, and exudative retinal detachment.1– 8 It usually occurs unilaterally in young males and, if untreated, can lead to total retinal detachment and secondary glaucoma, sometimes requiring enucleation.1– 8 There have been no reports that analyze statistically the

ELSEVIER SCIENCE INC. ALL

RIGHTS RESERVED.

0002-9394/01/$20.00 PII S0002-9394(01)00896-0

clinical factors predictive of poor visual outcome and enucleation. In a recent study of 150 cases, we outlined criteria for diagnosing Coats disease and for differentiating it from other causes of exudative retinopathy.8 In this communication, we use the same patient database to analyze results of treatment, propose a staging classification, and make recommendations regarding management of Coats disease.

PATIENTS AND METHODS RECORDS WERE REVIEWED ON PATIENTS WITH THE DIAG-

nosis of Coats disease from the computerized patient database of the Oncology Service at Wills Eye Hospital from January 1975 through December 1999. We defined Coats disease as idiopathic retinal telangiectasia and intraretinal or subretinal exudation without appreciable signs of vitreoretinal traction.8 Other causes of exudative retinopathy that were excluded from this study, as well as general data, referral diagnoses, and clinical features, were reported in a separate study of 150 patients.8 In this study, we review the methods and results of treatment for patients with Coats disease in 117 of those patients (124 eyes) who were managed by us and who had follow-up of 6 months or longer. The primary goal of treatment was to eradicate retinal telangiectasia to achieve resolution of intraretinal and subretinal exudation with the ultimate goal of preserving the globe and vision. The selection of management was based on the clinical findings and the experience of the senior authors (Drs J. Shields and C. Shields). Depending on the clinical situation, the selected management was observation, laser photocoagulation, cryotherapy, and surgical repair of retinal detachment, combined with laser photocoagulation or cryotherapy, or enucleation. Observation was generally recommended in either of two situations. First were patients with telangiectasias and little or no exudation and no impending threat to vision. Second were patients with a comfortable eye but total retinal detachment, in which there was no hope for useful vision. Photocoagulation to the telangiectasias was usually selected when there was exudation but no retinal detachment or a very shallow detachment. In the early part of this study, xenon arc photocoagulation was used and later argon laser was employed, originally with slit-lamp delivery and more recently with indirect ophthalmoscope delivery. Cryotherapy, with a double freeze-thaw method, was employed when the retinal detachment was shallow enough to allow approximation of a cryoprobe to the telangiectasias using scleral indentation. It was usually performed by a transconjunctival technique, but a small conjunctival incision was sometimes necessary to reach more posteriorly located telangiectasias. Surgical reattachment of the retina was often attempted VOL. 131, NO. 5

CLASSIFICATION

AND

when the detachment was so extensive that cryotherapy or photocoagulation would have been ineffective. In most earlier cases, the surgical approach involved performing a sclerotomy and draining subretinal fluid while simultaneously injecting saline into the anterior chamber or vitreous.9 More recently, a vitreoretinal surgical approach using vitrectomy and intravitreal infusion of saline or expandable gas was employed. Supplemental photocoagulation or cryotherapy was applied to the telangiectasias at the time of retinal reattachment, to eradicate them and create a chorioretinal adhesion. Enucleation was usually performed when the affected eye was blind and painful from secondary glaucoma. In those cases, there was no hope for vision using conservative methods of treatment. Patients were evaluated at 3-month to 4-month intervals, depending on the status of the disease, distance necessary for patient travel, and the degree of participation by the referring physician. The methods of initial management and the number and type of subsequent treatments were reviewed. The resolution of telangiectasia, retinal exudation, subretinal fluid, and retinal macrocysts and the frequency of late recurrence and its management were determined. Based on follow-up examinations, an assessment of the final ocular outcome (improved, stable, worse,) was made. The final visual acuity and reasons for poor visual acuity (20/200 or worse) were tabulated. Using univariate and multivariate analyses, the clinical findings were assessed to determine the factors predictive of poor visual outcome of 20/200 or worse. The complications of treatment and reasons for enucleation were determined. Using univariate and multivariate analyses the clinical factors predictive of enucleation were determined. Based on these data, a staging classification for Coats disease, applicable to treatment selection and ocular prognosis, was proposed (Figure 1).

RESULTS OF THE 150 PATIENTS (158 EYES) IN THE SERIES, 25 HAD

initial treatment elsewhere, and limited information was available on the techniques and results of management. There were 125 patients (132 eyes) managed exclusively by the authors of which 117 patients (124 eyes) had at least 6 months follow-up and were included in this analysis. The mean follow up was 55 months (median, 23 months; range, 6 months to 24 years). The methods of management at the time of diagnosis in the 124 eyes are shown in Table 1. Observation was recommended for 22 of the 124 eyes (18%), of which six had minimal, asymptomatic disease and 16 had advanced disease with total, bullous retinal detachment. Of the six cases with minimal disease followed with observation, five remained stable and one showed progression and required cryotherapy. Of the 16 eyes with advanced disease man-

MANAGEMENT

OF

COATS DISEASE

573

FIGURE 1. Examples of stages of Coats disease (see Table 11). (A) Stage 1. Retinal telangiectasia only. (B) Stage 2A. Telangiectasia and extrafoveal exudation. The exudation is confined to the equatorial region and the fovea is spared. (C) Stage 2B. Foveal exudation. (D) Stage 3A1. Subtotal retinal detachment inferiorly, sparing the fovea. (E) Stage 3A2. Subtotal retinal detachment extending beneath the fovea. (F) Stage 3B. Total exudative retinal detachment. Note the typical telangiectasias and the yellow-green color to the subretinal fluid. (G) Stage 4. Total retinal detachment behind the lens in eye with secondary glaucoma. (H) Stage 5. Advanced end stage disease with chronic inflammation, posterior synechia and cataract, secondary to longstanding retinal detachment.

574

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

MAY 2001

TABLE 1. Primary Management in 117 Consecutive Patients (124 Eyes) With Coats Disease Management

Number (%)

Observation Photocoagulation Cryotherapy Retinal detachment repair with subretinal fluid drainage and cryotherapy or photocoagulation Enucleation

22 (18%) 16 (13%) 52 (42%) 20 (17%) 14 (11%)

aged with observation, 13 remained stable with a blind but comfortable eye and three showed progression to painful secondary glaucoma and required enucleation. Laser photocoagulation was the initial treatment for 16 eyes (13%). Cryotherapy was employed in 52 eyes (42%). Surgical management for retinal detachment, combined with either laser photocoagulation or cryotherapy, was employed in 20 eyes (17%). Primary enucleation was performed in 14 eyes (11%). A second treatment was necessary in 60 eyes. This included cryotherapy in 34, laser photocoagulation in 11, combined photocoagulation and cryotherapy in four, and enucleation in four. The other seven eyes were treated with combinations of drainage of subretinal fluid, cryotherapy, and scleral bucking procedures. A third treatment was undertaken in 27 eyes. This included cryotherapy in 13, laser photocoagulation in four, combinations of drainage of subretinal fluid, cryotherapy, and scleral bucking procedures in eight, and enucleation in two cases. In all, 18 eyes (15%) were observed, 31 (25%) underwent laser photocoagulation, 99 (80%) underwent cryotherapy, and 35 (28%) had subretinal fluid drainage with or without scleral buckling, pars plana vitrectomy, laser photocoagulation, or cryotherapy. Of the 76 eyes treated with cryotherapy, one treatment session was necessary in 36 (47%), two sessions in 19 cases (25%), three sessions in 16 (21%), and four to five sessions were required in five (7%) to achieve successful obliteration of the telangiectasias. There were 103 eyes in which we were able to follow ophthalmoscopically the course of the telangiectasia and exudation after treatment (excluding those managed by primary enucleation, those not treated, and some with hazy ocular media). Of those 103 eyes, the telangiectasia resolved completely in 48 eyes (47%) and partially resolved in 55 (53%). The mean interval from initial treatment to complete resolution of telangiectasia was 15 months (median, 10 months; range, 2 to 123 months). If the retinal exudation was resolving satisfactorily, attempts to completely eradicate the telangiectasia were not undertaken, because the ultimate goal was to preserve vision and not necessarily to eradicate all telangiectasias The exudation resolved completely after treatment of the telangiectasias in 46 cases (45%; Figure 2). The mean VOL. 131, NO. 5

CLASSIFICATION

AND

time to resolution of the exudation was 17 months (median, 12 months; range, 1 to 42 months). If residual exudation was not causing visual impairment and the telangiectasias were controlled, then no further treatment was advised. In 88 eyes with retinal detachment that were treated and had adequate follow-up information, the detachment resolved completely in 50 (57%; Figure 3). The mean time from the first treatment to resolution of subretinal fluid was 9 months (median, 5 months; range, 1 to 43 months). In 38 cases (43%), residual subretinal fluid persisted after retinal detachment surgery. There were 15 eyes in which a retinal macrocyst was observed in an area of retinal detachment. The macrocyst resolved after treatment in five eyes, and it persisted in 10. There was eventual recurrence of telangiectasia and exudation in six of the 86 eyes (7%) after initial satisfactory control of telangiectasia and exudation. The mean time for recurrence was 10 years (median, 9 years; range, 4 to 14 years). The final anatomic outcome, as determined by improvement in the amount of telangiectasias, exudation, and subretinal fluid, was assessed in patients with follow-up of at least 6 months (Table 2). There was stabilization or improvement of the disease process in 76% and worsening in 8%. Overall, 20 patients (16%) required enucleation either initially (14 eyes) or in the course of management (six eyes). The final visual outcome was determined in 104 of the 124 eyes (Table 3). Of the 104 eyes, the visual acuity was 20/50 or better in 17 (16%), between 20/60 and 20/100 in eight (8%), from 20/200 to finger counting in 30 (29%), and hand motions to no light perception in 49 (47%), and 20 eyes were enucleated. The reasons for final vision of 20/200 or worse included various combinations of persistent retinal detachment, exudation, and fibrosis involving the fovea (Table 4). The subretinal exudation often transformed with time into dense fibrous tissue. The significant clinical factors predictive of poor visual acuity (20/200 or worse) are shown in Tables 5 and 6. Using multivariate analysis, the most significant factors related to poor visual outcome included non-Caucasian race, diffuse, postequatorial, and superior location of the telangiectasias and exudation, failure of exudation to resolve after treatment, and presence of retinal macrocysts (Table 6). The treatment complications are shown in Table 7. Of the 124 eyes, 86 were managed by interventional treatment other than enucleation. There were no serious complications of treatment in 78 of those eyes (91%). Vitreous hemorrhage occurred in three eyes (3%), and a retinal hole developed in two cases (2%) after cryotherapy. Although a mild transient retinal detachment (ablatio fugax) was observed frequently after cryotherapy, only two eyes (2%) had a permanent increase in the extent of retinal detachment. Enucleation was performed in 20 cases (16%). In 14 cases, it was performed at the time of the initial visit

MANAGEMENT

OF

COATS DISEASE

575

FIGURE 2. Resolution of foveal exudation after treatment of peripheral telangiectasia in a 17-year-old male. (Top left) Retinal telangiectasia located superotemporally. (Top right) Exudation extending into foveal area. The fovea is in lower left part of photograph. (Bottom left) Area of telangiectasia after 2 years, showing area of pigmentation. (Bottom right) Appearance of fovea 14 years after treatment, showing continued absence of the exudation. Patient was 31 years of age at the time of the photograph.

enucleated, and all showed retinal telangiectasia, total retinal detachment, and eosinophilic subretinal fluid containing lipid-laden macrophages and cholesterol clefts.4,5,10 In 12 cases, cytology of subretinal fluid drained through a posterior sclerotomy at the time of retinal detachment repair revealed characteristic cytologic features of Coats disease, with lipid-laden macrophages and cholesterol crystals. In six of the earlier cases, diagnostic fine needle aspiration biopsy of subretinal fluid11 demonstrated typical cytologic features of Coats disease. Based on our observations, a staging classification of Coats disease was devised as follows:

because of a total retinal detachment, usually associated with secondary glaucoma. In the other six cases, enucleation was elected after a period of follow-up with or without treatment. The reasons for enucleation in the 20 cases, shown in Table 8, included neovascular glaucoma in nine, severe angle closure glaucoma without iris neovascularization in four, blind painful eye without glaucoma in six, and parental concern about slight possibility of retinoblastoma in one blind eye (Table 8). The clinical risk factors predictive of enucleation were identified using univariate (Table 9) and multivariate (Table 10) analyses. With multivariate analysis (Table 10), the most significant factors were elevated intraocular pressure, and iris neovascularization, and heterochromia, usually secondary to iris neovascularization. Microscopic verification of Coats disease was obtained in 38 of the 124 eyes. As mentioned, 20 eyes were 576

AMERICAN JOURNAL

Stage 1: Retinal telangiectasia only Stage 2: Telangiectasia and exudation A. Extrafoveal exudation B. Foveal exudation OF

OPHTHALMOLOGY

MAY 2001

FIGURE 3. Resolution of foveal retinal detachment after treatment of reinal telangiectasia in an 8-year-old male. (Top left) Retinal telangiectasia and exudation near equator nasally. (Top right) Macular area, showing mild exudation and a serous detachment of fovea. (Bottom left) Appearance immediately after laser treatment to the telangiectasia. (Bottom right) Macular area 9 months after treatment, showing resolution of the retinal detachment and only minimal residual exudation.

Stage 3: Exudative retinal detachment A. Subtotal detachment 1. Extrafoveal 2. Foveal B. Total retinal detachment Stage 4: Total retinal detachment and glaucoma Stage 5: Advanced end-stage disease Stage 1 is characterized by telangiectasia only (Figure 1, A), stage 2 by telangiectasia and exudation (Figure 1, B and C), stage 3 by exudative retinal detachment (Figures 1, D, E, and F), stage 4 by total retinal detachment with glaucoma (Figure 1, G), and stage 5 by advanced end-stage changes (Figure 1, H). Stage 5 disease was defined as a blind, nonpainful eye with a total retinal detachment, often with cataract and phthisis bulbi, Of the 124 eyes reviewed, there was one eye in stage 1, 10 in stage 2A, VOL. 131, NO. 5

CLASSIFICATION

AND

seven in 2B, 25 in 3A1, 24 in 3A2, 37 in 3B, 18 in stage 4, and three in stage 5 (Table 11). The treatment employed varied with the stage of the disease. In stages 1 and 5 the most common management was observation (100%). Cryotherapy was most often used in stage 2A (50%), 2B (100%), 3A1 (56%), 3A2 (65%), and 3B (34%). Photocoagulation was the selected initial treatment in 16 (13%) of our cases, mainly in stages 2 and 3A. Surgical repair of retinal detachment was employed mainly in eyes with stage 3B disease. The visual outcome was generally favorable in stages 1 and 2A and was worse in stage 2B and higher. The chances of enucleation were greatest in stage 4 because of ocular pain secondary to glaucoma. (Table 11). In stage 4, enucleation was done initially in 14 eyes (78%). The other six eyes (22%) that came to enucleation were initially classified as stage 3

MANAGEMENT

OF

COATS DISEASE

577

TABLE 2. Final Ocular Outcome in 117 Patients With Coats Disease (124 Eyes) With Follow-up of at Least 6 Months Final Outcome

Number (%)

Improved Stable Worse Enucleation

59 (48%) 35 (28%) 10 (8%) 20 (16%)

TABLE 4. Reasons for Final Visual Acuity of 20/200 or Worse (79 Eyes) With Coats Disease

Number (%)

20/50 or better 20/60–20/100 20/200–CF HM–NLP Enucleated

17 (14%) 8 (6%) 30 (24%) 49 (40%) 20 (16%)

disease, but later showed progression, with blindness and discomfort.

DISCUSSION SEVERAL AUTHORS HAVE REPORTED EXPERIENCE WITH

management of Coats disease.12–29 Methods have included diathermy, xenon photocoagulation, argon laser photocoagulation, cryotherapy, and techniques of surgical reattachment of the retina. The main goal of treatment should be to eradicate the telangiectasias to facilitate resolution of exudation and salvage as much vision as possible. Prior publications on management of Coats disease have been retrospective studies, often on small groups of patients.12–29 In addition, some series6,7 have included conditions that do not meet diagnostic criteria for Coats disease, as described originally by Coats1 and more recently by others.2,3,8 Many of the conditions that have been misinterpreted as Coats disease are listed in our prior report on clinical variations of that disease.8 Because our patients were managed over a long span of time during which treatment modalities were changing, it was not possible to conduct a prospective study comparing results of treatment methods. Because Coats disease is relatively uncommon and the degree of involvement varies from case to case, a randomized, prospective study of treatment would probably not be feasible. Our study is also retrospective, and AMERICAN JOURNAL

Subfoveal fluid Foveal exudation Subfoveal fibrosis Subfoveal hemorrhage Macular edema Epiretinal membrane Optic atrophy

37 (47%) 9 (11%) 23 (29%) 1 (1%) 6 (8%) 2 (2%) 1 (1%)

therefore, the results must be interpreted with caution. However, the relatively large number of patients in our series gave us the unique opportunity to analyze statistically the risk factors for poor visual outcome and for enucleation. Our observations have allowed us to develop insight into the therapeutic options for Coats disease, which includes observation, laser photocoagulation, cryotherapy, surgical approaches to reattach the retina, or combinations thereof. It is important to realize that patients undergoing treatment need close follow-up and may require additional treatments, depending on clinical developments. There are two diverse situations in Coats disease where observation is warranted. First is in eyes with stage 1 and 2A disease with mild telangiectasia and little exudation, particularly in patients older than 15 years. In older children and young adults, Coats disease is generally less aggressive and there is less likelihood of progressive exudation and retinal detachment. However, treatment should be considered if progression is documented. There were six such cases in our series, and only one showed progression of the disease during the course of follow-up. In some older children and young adults, spontaneous regression of the telangiectasias has been documented.22 A second reason to observe Coats disease is in some eyes with stages 3B and 5 disease, where the blind eye is comfortable, but there is no hope for useful vision. In our study, 13 of 16 eyes (81%) with such advanced disease remained stable during the period of observation. Laser photocoagulation can be used in selected cases of Coats disease.14 –21 Several decades ago, xenon photocoagulation was the only available method.13–16 Because it caused more retinal damage and vitreoretinal traction, xenon photocoagulation has been replaced by argon laser photocoagulation today. More recently, we have employed the indirect ophthalmoscope delivery system, because it is more readily used in the operating room in young children.24,30 Photocoagulation to the telangiectasia was the initial treatment in 16 (13%) of our cases (Table 1). In our experience, laser photocoagulation was most successful

CF ⫽ counting fingers; HM ⫽ hand motions; NLP ⫽ no light perception. Data on final visual acuity was available in 104 eyes, with 20 having been enucleated.

578

Number (%)

Percents may not equal 100 because of rounding.

TABLE 3. Final Visual Acuity (124 Eyes) With Coats Disease Final visual acuity

Reason

OF

OPHTHALMOLOGY

MAY 2001

TABLE 5. Univariate Analysis of Risk Factors for Poor Visual Outcome (20/200 or Worse) in 117 Consecutive Patients (124 Eyes) With Coats Disease Good Visual Outcome (n ⫽ 25)

Variable

Non-Caucasian race Retinal zone mainly affected Equator to macula Diffuse* Retinal telangiectasia Affecting the superior retinal sector† Unresolved retinal telangiectasia Retinal exudation Diffuse retinal exudation* Unresolved retinal exudation Subretinal fluid Unresolved subretinal fluid Presence of retinal macrocyst Late onset recurrence

Poor Visual Outcome (n ⫽ 99)

P Value

Relative Risk

4

16

.018

2.01

1.13–3.58

3 10

6 89

.007 .004

5.93 5.81

1.63–21.61 1.78–19.02

2 11

5 46

.017 .001

0.26 2.20

0.09–0.79 1.37–3.53

10 7

79 50

.041 .001

3.42 0.45

1.05–11.10 0.28–0.72

3 1 3

35 14 3

.026 .017 .009

0.57 1.60 0.21

0.35–0.94 1.09–2.35 0.06–0.67

95% Confidence Interval

*Diffuse ⫽ involvement of retina from ora serrata to foveal region. Superior retinal sector ⫽ between 10 and 2 o’clock.

†

TABLE 6. Multivariate Analysis of Risk Factors for Poor Visual Outcome (20/200 or Worse) in 117 Consecutive Patients (124 Eyes) With Coats Disease Good Visual Outcome (n ⫽ 25)

Variable

Non-Caucasian race Retinal zone predominantly affected Equator to macula Diffuse Retinal telangiectasia Affecting the superior retinal sector* Retinal exudation Unresolved retinal exudation Presence of retinal macrocyst

Poor Visual Outcome (n ⫽ 99)

P Value

Relative Risk

95% Confidence Interval

4

16

.04

1.92

1.05–3.52

3 10

6 89

.001 .01

12.57 6.20

2.79–56.55 1.52–25.31

2

5

.04

0.34

0.12–0.95

7 1

50 14

.02 .01

0.55 1.82

0.33–0.90 1.16–2.87

*Superior retinal sector ⫽ between 10 and 2 o’clock. † Diffuse ⫽ involvement of retina from ora serrata to foveal region.

when there were telangiectasias without retinal detachment (stage 2). Cryotherapy can also be used in such cases. The patient should be reevaluated 3 months after treatment and, if vascular changes persist and exudation is not resolving, a second treatment should be given. With time, remote exudation in the macular area will resolve if the peripheral retinal vascular abnormalities are obliterated. In some instances, more than three treatments may be necessary. It is important to wait at least 3 months before considering additional laser photocoagulation, because resolution of the exudation is generally very slow. Cryotherapy is also an important modality in the treatment of Coats disease.17–21 It was the initial treatment in 42% of our patients. In our experience, it is the preferred initial method when there are peripheral telangiectasias associated with extensive exudation or subtotal retinal VOL. 131, NO. 5

CLASSIFICATION

AND

detachment (stage 3A). We have learned that cryotherapy can be employed even in cases of relatively high detachment (stage 3B), provided the peripheral telangiectasias can be reached with scleral indentation with a cryoprobe. The preferred method of cryotherapy is a double freezethaw technique directly to the vascular abnormalities. Excessive cryotherapy can induce a transient increase in the extent of retinal detachment (ablatio fugax). Therefore, in cases with diffuse telangiectasias involving all quadrants, it is generally advisable to treat two quadrants only and to treat the other quadrants in 4 weeks. As in the case of laser photocoagulation, a second treatment to the same area should be withheld for approximately 3 months to allow time for exudation to resolve and to avoid excessive treatment. Most authorities agree that the visual prognosis is poor

MANAGEMENT

OF

COATS DISEASE

579

TABLE 7. Complications of Treatment in 86 Eyes With Coats Disease Managed by Interventional Treatment in 117 Patients (124 Treated Eyes)*

TABLE 8. Management of Coats Disease in 117 Consecutive Patients: Reason for Enucleation in 20 Patients*

Complications

Number (%)

Reason for Enucleation

Number (%)

Vitreous hemorrhage Retinal hole Increase in subretinal fluid Rhegmatogenous retinal detachment No complication

3 (3%) 2 (2%) 2 (2%) 1 (1%) 78 (91%)

Acute neovascular glaucoma Angle-closure glaucoma Suspected retinoblastoma Painful blind eye*

9 (45%) 4 (20%) 1 (5%) 6 (30%)

*All enucleated eyes were experiencing some pain. *Enucleation was performed in 20 eyes and 18 eyes were observed, without interventional treatment. Those were excluded from this calculation.

then a favorable visual outcome can be expected (Table 11). Even dense macular exudation may gradually resolve after successful eradication of the peripheral telangiectasias. However, the presence of thick foveal exudation (stage 2B and above) usually portends a worse prognosis, particularly if a fibroglial nodule is present in the fovea. Patients with a total retinal detachment usually have little or no useful vision in the affected eye despite treatment. The fact that 47% of our patients had a final visual acuity in the hand motion to no light perception range probably reflects the fact that our practice is oriented largely toward ocular tumors, and we are often referred young children with a total retinal detachment in which exophytic retinoblastoma is a diagnostic consideration. Younger children with Coats disease generally have a more aggressive clinical course and often have a total retinal detachment by the time the disease is diagnosed.8,29 Hence, a poor visual outcome is usually anticipated in such cases. In other practices oriented toward retinal vascular diseases, it is possible that there would be more referrals for earlier stages of Coats disease and a better visual prognosis could be anticipated. The reasons for a poor visual outcome in patients with Coats disease have not been specifically addressed in most prior publications. We found that persistent subretinal fluid and late subretinal fibrosis in the foveal area portended a worse visual outcome (Table 4). Even with successful reattachment of the retina, the subretinal exudation often leads to irreversible subretinal fibrosis and retinal degeneration. In the multivariate analysis for poor visual outcome (Table 6), it was not surprising that there were worse visual results when the telangiectasias and exudation were more extensive and involved the foveal area. There are few serious complications of the aforementioned treatments. No major complications were encountered in 91% of the cases. The occasional complications (Table 7) had little visual impact, because many affected eyes had advanced disease with little or no visual potential. In some cases, enucleation for Coats disease is warranted. Enucleation was necessary as initial or secondary treatment in 16% of our patients. As mentioned earlier,

in patients with total, bullous retinal detachment secondary to Coats disease.26,29 However, in selected cases, an attempt to repair the retinal detachment surgically may be justified to prevent the development of neovascular glaucoma. In 1988, Silodor and associates23 reported 13 eyes with total bullous retinal detachment. Of six such eyes that were not treated, four developed painful neovascular glaucoma. In contrast, no neovascular glaucoma developed in any of the seven eyes that had drainage of subretinal fluid, intraocular infusion to flatten the retina, and supplemental photocoagulation or cryotherapy. More recently, Yoshizumi and associates27 have advocated vitrectomy, internal drainage of subretinal fluid, intraocular diathermy, and intravitreal gas or silicone oil injection as a means of salvaging such eyes. Various combinations of these more aggressive methods were employed as primary treatment in 17% of our 124 patients, mainly when there was a total bullous retinal detachment without secondary elevated intraocular pressure (stage 3B). This option should be considered when there is a reasonable chance of reattaching the retina and preventing neovascular glaucoma, even though the visual outcome is expected to be poor. The concept of late recurrence of Coats disease has not been adequately emphasized in the literature. We found that there was recurrence of the telangiectasia and retinal exudation in 7% of cases after a mean of 10 years after successful treatment. This stresses the importance of longterm follow-up of affected patients. Based on our statistical observations, we believe that the ophthalmologist can make a general prediction as to the chances of retaining the eye and preserving vision in patients with Coats disease. From an anatomic standpoint, improvement or stabilization of the disease was achieved in almost 76% of our cases (Table 2). Therefore, the majority of patients respond favorably to treatment, but approximately 25% will become worse or require enucleation. The visual outcome in patients with Coats disease varies considerably (Table 3). Our observations suggest that if the patient does not have macular exudation or extensive retinal detachment, such as in stages 1 and 2A disease, 580

AMERICAN JOURNAL

OF

OPHTHALMOLOGY

MAY 2001

TABLE 9. Management of Coats Disease in 117 Consecutive Patients (124 Eyes): Univariate Analysis of Risk Factors for Enucleation Variable

No Enucleation (n ⫽ 104)

Enucleation (n ⫽ 20)

P Value

Relative Risk

95% Confidence Interval

Visual acuity 20/60 or worse Leukocoria Heterochromia Mean ocular pressure (mm Hg) Corneal edema Shallow anterior chamber Neovascularization of iris Retinal telangiectasia Mean number of sectors affected Mean number of clock hours affected Retinal exudation Macular exudation Diffuse retinal exudation* Mean number of sectors affected Mean number of clock hours affected Subretinal fluid Diffuse subretinal fluid* Mean number of clock hours affected Mean percent affected

92 23 3 17 0 3 3

20 8 6 35 5 4 9

.04 .02 ⬍.001 ⬍.001 ⬍.001 .001 ⬍.001

6.47 4.75 25.36 1.11 68.28 6.12 13.65

1.06–40.01 1.23–18.37 6.28–102.48 1.08–1.15 16.76–278.09 2.02–18.58 5.48–34.01

2 6

3 8

.03 .03

1.57 1.14

1.04–2.39 1.01–1.29

84 51 3 9

20 18 4 12

.007 .001 .01 .02

NA 7.30 2.53 1.32

NA 1.69–31.60 1.24–5.16 1.04–1.68

58 9 65

20 12 93

⬍.001 .04 .02

NA 1.61 1.03

NA 1.02–2.52 1.01–1.05

NA ⫽ relative risk and 95% confidence interval could not be calculated. *Diffuse ⫽ involvement of retina from ora serrata to foveal region.

TABLE 10. Management of Coats Disease in 117 Consecutive Patients (124 Eyes): Multivariate Analysis of Risk Factors for Enucleation Variable

No Enucleation (n ⫽ 104)

Enucleation (n ⫽ 20)

P Value

Relative Risk

95% Confidence Interval

Heterochromia Mean intraocular pressure (mm Hg) Neovascularization of iris

3 17 3

6 35 9

⬍.001 ⬍.001 ⬍.001

12.86 1.10 13.65

3.18–52.02 1.06–1.15 5.48–34.01

younger children with Coats disease often present with acute glaucoma, characterized by pain, nausea, vomiting, loss of appetite, and lethargy. When this occurs, we believe that palliative enucleation is justified, to spare the child chronic pain and oral and topical medication. We do not usually recommend diagnostic fine needle aspiration biopsy in such advanced cases. If enucleation is anticipated, then needle biopsy is an unnecessary additional procedure. In the multivariate analysis of risk factors for enucleation, we found that iris neovascularization with secondary glaucoma were most important (Table 10). Historically, patients with Coats disease often underwent enucleation because retinoblastoma was a serious diagnostic consideration.31,32 Today, however, features that differentiate Coats disease retinoblastoma are better known, and enucleation because of erroneous diagnosis of retinoblastoma is less common.8,33,34 The proposed classification of Coats disease can be helpful for selecting treatment and predicting the ocular and visual VOL. 131, NO. 5

CLASSIFICATION

AND

outcomes (Table 11). Patients with stage 1 disease (telangiectasia only) can be managed by either periodic observation or laser photocoagulation. In stage 1 disease, there is high probability that the eye can be salvaged and the visual prognosis is usually favorable. However, stage 1 disease is uncommon in a clinical practice, and Coats disease is usually more advanced at the time of diagnosis. Patients with stage 2 disease (telangiectasia and exudation) are generally best managed by laser photocoagulation or cryotherapy, depending on the extent of the disease and the preference of the ophthalmologist. If the exudation is limited to one quadrant or located nasally, a reasonably good visual outcome can be anticipated. In stage 2A, the visual prognosis is generally good, because the fovea is not involved by exudation. Eyes with stage 2B are usually salvaged and the visual prognosis is fairly good if the foveal exudation is not advanced (Figure 2). As mentioned earlier, a dense yellow gray nodule centered within the foveal exudation usually portends a worse visual outcome.

MANAGEMENT

OF

COATS DISEASE

581

TABLE 11. Management of 117 Patients (124 Eyes) With Coats Disease: Staging Classification, Visual Outcome, and Frequency of Enucleation Management Modality*

Stage

Staging Criteria

Stage 1 Stage 2 2A 2B Stage 3 3A 3A (i) 3A (ii) 3B Stage 4 Stage 5

RT T⫹Ex Extrafoveal Ex Foveal Ex T⫹Ex⫹RD Subtotal RD Extrafoveal RD Foveal RD Total RD T⫹Ex⫹Total RD⫹Gl End-stage disease

Outcome

Observation, Number (%)

PC, Number (%)

Cryotherapy, Number (%)

RD Repair, Number (%)

Poor VA, Number (%)

Enucleation, Number (%)

1 (%)

1 (100%)

0

0

0

0

0

10 (8%) 7 (6%)

4 (40%) 0

1 (10%) 3 (43%)

5 (50%) 7 (100%)

0 0

3 (30%) 6 (86%)

0 0

24 (19%) 24 (19%) 37 (30%) 18 (15%) 3 (2%)

4 (17%) 2 (8%) 4 (10%) 0 3 (100%)

10 (42%) 12 (50%) 4 (10%) 0 0

2 (8%) 2 (8%) 18 (49%) 4 (22%) 0

16 (67%) 16 (67%) 31 (84%) 4 (22%) 3 (100%)

1 (4%) 1 (4%) 4 (11%) 14 (78%) 0

Number (%)

14 (58%) 15 (63%) 31 (84%) 4 (22%) 0

Ex ⫽ exudation; Gl ⫽ glaucoma; PC ⫽ photocoagulation; RD ⫽ retinal detachment; RD repair ⫽ subretinal fluid drainage with or without scleral buckling, pars plana vitrectomy, laser, photocoagulation, or cryotherapy; RT ⫽ retinal telangiectasia; VA ⫽ visual acuity 20/200 or worse. *Includes all primary and subsequent management.

detachment with glaucoma) are often best managed by enucleation to relieve the severe ocular pain. Patients with stage 5 disease generally have a blind, but comfortable, eye and require no aggressive treatment. In summary, we have reviewed our experience with management of Coats disease, determined statistical factors predictive of poor visual outcome and enucleation, and proposed a useful classification of this condition. Treatment should be directed toward obliterating the telangiectasias, thus facilitating resolution of the exudation and retinal detachment. In the early stages, this can be best achieved with laser photocoagulation or cryotherapy. More advanced cases may require surgical techniques of retinal reattachment, combined with photocoagulation or cryotherapy. Advanced cases with secondary glaucoma often require enucleation. The proposed staging classification should assist the ophthalmologist in selecting treatment and in predicting the prognosis for salvaging the eye and preserving vision. Carefully selected treatment can anatomically stabilize or improve the eye with Coats disease in 76% of cases. However, poor visual outcome of 20/200 or worse commonly results. Patients who present with stages 1 to 3 Coats disease have the best ocular prognosis. We believe that earlier diagnosis and prompt treatment will result in better visual outcome and fewer enucleations for Coats disease in the future.

TABLE 12. Staging Classification of Coats Disease, Visual Outcome, and Frequency of Enucleation

Stage

Stage 1: Retinal telangiectasia only Stage 2: Telangiectasia and exudation A. Extrafoveal exudation B. Foveal exudation Stage 3: Exudative retinal detachment A. Subtotal detachment 1. Extrafoveal 2. Foveal B. Total detachment Stage 4: Total detachment; glaucoma Stage 5: Advanced end-stage disease

Poor Visual Outcome*

Enucleation Outcome

(n ⫽ 1) 0%

0%

(n ⫽ 10) 30% (n ⫽ 7) 86%

0% 0%

(n ⫽ 25) 70% (n ⫽ 23) 70% (n ⫽ 37) 94% (n ⫽ 18) 100% (n ⫽ 3) 100%

8% 0% 11% 78% 0%

*Visual acuity 20/200 or worse (excluding enucleated eyes).

Patients with stage 3A disease (subtotal retinal detachment) can generally be managed by photocoagulation or cryotherapy. Even if the retinal detachment involves the fovea, it will resolve when the telangiectasias are eradicated (Figure 3). Laser photocoagulation is less effective in areas of retinal detachment, and cryotherapy is often preferable in such instances. Patients with stage 3B (total retinal detachment) can be managed with cryotherapy if the detachment is shallow but may require an attempt at surgical reattachment if the detachment is bullous and immediately posterior to the lens. Patients who present with stage 4 disease (total retinal 582

AMERICAN JOURNAL

REFERENCES 1. Coats G. Forms of retinal diseases with massive exudation. Roy Lond Ophthalmol Hosp Rep 1908;17:440 –525. 2. Mandava N, Yannuzzi LA. Coats’ disease. In: Guyer DR, OF

OPHTHALMOLOGY

MAY 2001

3. 4. 5. 6. 7. 8.

9. 10. 11.

12. 13. 14. 15. 16. 17. 18.

Yannuzzi LA, Chang S, Shields JA, Green WR, editors. Retina-vitreous-macula. Philadelphia: WB Saunders Company, 1999:390 –397. Haller JA. Coats’ disease. In: Ryan SJ, editor. Retina, edition 2, volume 2. St. Louis: CV Mosby Company, 1994:1453– 1460. Shields JA, Shields CL. Coats’ disease. In: Shields JA, Shields CL, editors. Intraocular tumors. A text and atlas. Philadelphia: WB Saunders, 1992:356 –358. Shields JA, Shields CL. Coats’ disease. In: Shields JA, Shields CL, editors. Atlas of intraocular tumors. Philadelphia: Lippincott, Williams and Wilkins, 1999:234 –236. Woods AC, Duke JR. Coats’ disease. I. Review of the literature, diagnostic criteria, clinical findings, and plasma lipid studies. Br J Ophthalmol 1963;47:385– 412. Campbell FP. Coats’ disease and congenital vascular retinopathy. Trans Am Ophthalmol Soc 1977;74:365– 424. Shields JA, Shields CL, Honavar S, Demirci H. Clinical variations and complications of Coats disease in 150 cases: The 2000 Sanford Gifford Memorial Lecture. Am J Ophthalmol 2001;131:561–571. Silodor SW, Augsburger JJ, Shields JA, Tasman WS. Natural history and management of advanced Coats’ disease. Ophthalmic Surg 1988;19:89 –93. Eagle RC Jr. Coats’ disease. In: Eagle RC, Jr, editor. Eye pathology. Philadelphia: WB Saunders, 1999:214 –215. Shields JA, Shields CL, Ehya H, Eagle RC Jr, DePotter P. Fine needle aspiration biopsy of suspected intraocular tumors. The 1992 Urwick Lecture. Ophthalmology 1993;100: 1677–1684. Reese AB. Telangiectasis of the retina and Coats’ disease. Am J Ophthalmol 1956;42:1– 8. Morales AG. Coats’ disease. Natural history and results of treatment. Am J Ophthalmol 1965;60:855– 865. McGrand JC. Photocoagulation in Coats’ disease. Trans Ophthalmol Soc UK 1970;90:47–56. Harris GS. Coats’ disease, diagnosis and treatment. Can J Ophthalmol 1970;5:311–320. Chisholm IA, Foulds WS, Christison D. Investigation and therapy of Coats’ disease. Trans Ophthalmol Soc UK 1974; 94:335–341. Egerer I, Tasman WS, Tomer TL. Coats’ disease. Arch Ophthalmol 1974;92:109 –112. Theodossiadis GP. Some clinical, fluorescein-angiographic, and therapeutic aspects of Coats’ disease. J Ped Ophthalmol Strabism 1974;16:257–262.

19. Spitznas M, Joussen M, Wessing A. Treatment of Coats’ disease with photocoagulation. Albrecht Von Graefes Arch Klin Exp Ophthalmol 1976;199:31–37. 20. Ridley ME, Shields JA, Brown GC, Tasman W. Coats’ disease. Evaluation of management. Ophthalmology 1982; 89:1381–1387. 21. Tarkkanen A, Laatikainen L. Coats’ disease: clinical, angiographic, histopathological findings and clinical management. Br J Ophthalmol 1983;67:766 –776. 22. Deutsch TA, Robb MF, Jampol LM. Spontaneous regression of retinal lesions in Coats’ disease. Can J Ophthalmol 1982;17:169 –172. 23. Silodor SW, Augsburger JJ, Shields JA, Tasman W. Natural history and management of advanced Coats’ disease. Ophthalmic Surg 1988;19:89 –93. 24. Sneed SR, Blodi CF, Pulido JS. Treatment of Coats’ disease with the binocular indirect argon laser photocoagulator. Arch Ophthalmol 1989;107:789 –790. 25. Pauleikhoff F, Wessing A. Long term results of the treatment of Coats’ disease. Fortschritte der Ophthalmologie 1989;86: 451– 455. 26. Haik BG. Advanced Coats’ disease. Trans Am Ophthalmol Soc 1991;89:371– 476. 27. Yoshizumi MO, Kreiger AE, Lewis H, et al. Vitrectomy techniques in late-stage Coats’-like exudative retinal detachment. Doc Ophthalmol 1995;90:387–394. 28. Budning AS, Heon E, Gallie BL. Visual prognosis of Coats’ disease. J Am Assoc Pediatr Ophthalmol Strabism 1998;2: 356 –359. 29. Char DH. Coats’ syndrome: long term follow up. Br J Ophthalmol 2000;84:37–39. 30. Friberg TR. Clinical experience with a binocular indirect ophthalmoscope laser delivery system. Retina 1987;7:28 –31. 31. Kogan L, Boniuk M. Causes for enucleation in childhood with special reference to pseudogliomas and unsuspected retinoblastoma. Int Ophthalmol Clin 1962;2:507–524. 32. Howard GM, Ellsworth RM. Differential diagnosis of retinoblastoma. A statistical survey of 500 children. I. Relative frequency of the lesions which simulate retinoblastoma. Am J Ophthalmol 1965;60:610 – 618. 33. Shields JA, Shields CL, Parsons HM. Review. Differential diagnosis of retinoblastoma. Retina, 1991;11:232–243. 34. Shields JA, Parsons HM, Shields CL, Shah P. Lesions simulating retinoblastoma. J Pediatr Ophthalmol Strabismus 1991;28:338 –340.

The full-text of AJO is now available online at www.ajo.com. Authors Interactive威, currently available in limited form, is undergoing an upgrade.

VOL. 131, NO. 5

CLASSIFICATION

AND

MANAGEMENT

OF

COATS DISEASE

583