Indocyanine TT of Chronic CRSC

Abstract

Purpose. To evaluate the potential benefit and complicationsof indocyanine green-mediated photothrombosis (IMP) inthe management of patients with persistent central serouschorioretinopathy (CSC).

Methods. Interventional noncomparative case series. Elevenpatients with CSC presenting with persistent subretinal fluidin optical coherence tomography (OCT) four months afterpresentation and decrease in visual acuity (VA) were sub-mitted to a single IMP session with 2mg/kg body weight ICGand application of 5.6W/cm2 light at 810nm. A continuousfollow-up was provided with best-corrected ETDRS VAassessment, and angiographic and OCT documentation 72hours before and at 2 days, 1 and 2 weeks, 1, 3, 6, and 12months after treatment.

Results. Pretreatment VA levels ranged from 20/32 - 1 to20/100 (mean, 20/63 + 2 [logMAR equivalent, 0.460 0.155]); post treatment levels ranged from 20/25 - 2 to 20/20(mean, 20/20 - 2 [logMAR equivalent, 0.038 0.048]). Tenout of eleven patients presented with VA levels of 20/25 2 weeks after treatment; the mean logMAR VA change of0.345 at that time was statistically significant (p < 0.05,Friedman test). OCT disclosed resolution of persistent sub-retinal fluid in all eyes. No recurrence was observed after 12 months of follow-up. Complications included transientretinal whitening in two patients, and associated occlusion ofretinal capillaries in one.

Conclusions. Photothrombosis using low-intensity 810nmlight to direct laser energy continuously at the active leakagesites after intravenous ICG infusion induced rapid VA recov-ery in patients with persistent CSC; accordingly, restoration

of the macular architecture was evidenced on OCT, and norecurrence was noted 12 months after IMP.

Keywords: hyperthermia; indocyanine green; laser; optical coherence tomography; photodynamic therapy; thermotherapy

Introduction

Central serous chorioretinopathy (CSC) is a disorder ofunknown origin characterized by a serous detachment of the sensory retina affecting the macular region of healthyyoung adults, generally between the ages of 20 and 50years.1,2 Symptoms related to the disease such as mild lossof central vision, micropsia, metamorphopsia, and impaireddark adaptation are generally transient and resolve sponta-neously in 3 to 4 months.24 Fluorescein angiography demon-strates focal leaks at the level of the retinal pigmentepithelium whereas areas of choroidal hyperpermeability are identified in indocyanine green (ICG) angiography.5

Although a good visual outcome is generally expected,several studies suggest that the presence of persistentpigment epithelial detachment and/or subretinal fluidobserved in some patients with CSC is related with reducedfinal visual acuity (VA).68

Received: August 25, 2002Accepted: November 18, 2002

Correspondence: Rogrio Alves Costa, MD, Rua Itlia, 1905, Apto 74, Araraquara, SP 14801-350, Brazil. Tel./Fax: 55 (16) 3331 2197, E-mail: [email protected]

Indocyanine green-mediated photothrombosis as a new technique

of treatment for persistent central serous chorioretinopathy

Rogrio A. Costa1,2, Luciana Scapucin1, Nilva S. Moraes1, Daniela Calucci1, Luiz A. Melo Jr.1, Jos A. Cardillo1,2 and Michel E. Farah1

1Instituto da Viso IPEPO, Department of Ophthalmology, Federal University of So Paulo, Brazil; 2Consultores de Retinae Associados, Araraquara, So Paulo, Brazil

Current Eye Research 0271-3683/02/2505-287$16.002002, Vol. 25, No. 5, pp. 287297 Swets & Zeitlinger

Drs. Costa, Cardillo, and Farah are consultants to Akorn, Inc., a United States manufacturer of indocyanine green. Any kind of financial support was herein provided by the company sinceconsulting agreement has just been settled.

288 R.A. Costa et al.

Investigators have been seeking to abbreviate the durationof symptoms in patients with CSC by treating the activeleakage site (ALS) detected on fluorescein angiography withlaser application of different wavelengths and light inten-sity.919 The general consensus from these studies is that laserapplication shortens the course of the disease with no relatedcomplication; however, some authors reported significantloss of contrast sensitivity and iatrogenically producedchoroidal neovascularization after laser treatments in eyeswith CSC.10,2022

ICG-mediated photothrombosis is a novel, noninva-sive laser-dye modality used to achieve selective vascularocclusion with minimal or no damage to adjacent neuralstructures.23 Recently, the use of such modality for the management of choroidal neovascularization in patients with age-related macular degeneration achieved selectivelesion regression, resolution of serous retinal detachment andmacular edema, and improvement in vision.24 The therapeu-tic effect are thought to arise rather from the photochemicalreactions between pathologic areas with increased ICGuptake and laser rather than the heat generated after RPElight absorption, since no angiographic damage is commonlydetected in the normal tissues involved in the treatment spot(Costa et al. Vitreous Society Annual Meeting, Puerto Rico2001). Based on these results, coupled with the fact that ICGuptake is increased in areas of ALS in CSC patients, ICG-mediated photothrombosis was applied in a prospective,interventional case series to eleven patients with persistentCSC. All patients received a complete evaluation before andafter a single treatment session using a standardized regimenand were seen at regular intervals with documentation ac-cording to a defined protocol. The angiographic appearance,as well as best-corrected VA, and exudative changes in OCTevaluation were documented at initial presentation, 72 hoursprior to treatment (baseline evaluation) and during follow-upfor as long as 1 year.

Patients and methods

The clinical study protocol was approved by the local In-stitutional Review Board before patient enrollment at theRetina & Vitreous section of the Ophthalmology Departmentat the Federal University of So Paulo began, and all partici-pants gave written informed consent before participating inthe study. All fluorescein and ICG angiography studies andOCT evaluations were performed by an experienced certifiedophthalmic technician (DC). All treatments were performedby a single retinal specialist (RAC).

Patient selection

The patient enrollment period was from January through July2001. Candidates were patients with CSC with angiographicidentification of areas of ALS, and OCT evidence of serousretinal detachment involving the macular region. The indica-

tion for the therapeutic intervention was based on the pres-ence of macula-involving subretinal fluid on OCT evaluationas well as the best-corrected ETDRS VA of 20/32 or less fourmonths after initial presentation. Patients were not offeredenrollment in the study if they had drusen, peripapillarychanges with atrophic or pigmented punched out chori-oretinal lesions, polypoidal lesions, uveitis, or any other oph-thalmic disorder that might affect visual function. Patientswere also excluded if they (1) were unable to cooperate forthe laser procedure; (2) were unable to comprehend thenature of the procedure or the study to give an informedconsent; (3) had an allergy to fluorescein, ICG, or iodine; (4)had previously undergone laser photocoagulation; or (5) hadopacity that obstructed laser access to the macular area. Allparticipants had to be available for follow-up visits at regularintervals for up to 1 year after the treatment. Evaluation 4months after initial presentation was considered as the base-line evaluation for eleven patients who met the eligibility cri-teria for the study. They were informed orally and in writingof the potential benefits and risks of the procedure and allsigned a written form stating that they understood and con-sented to ICG-mediated photothrombosis.

ICG-mediated photothrombosis (IMP)

IMP was performed using ICG (IC-Green, Akorn, Inc.,Buffalo Grove, USA) as a sensitizer. The drug at a dose of 2mg/Kg body weight was dissolved in 4ml of distilled waterand administered IV as follows: a loading dose of 2.0ml ofthe ICG solution was infused as a bolus, followed immedi-ately by a 5.0-ml saline flush; twenty minutes after theloading dose, a second IV injection of the remaining ICGsolution (2ml) was administered, followed by a 5.0-ml salineflush. A diode laser (Trimode-L, OPTO, So Carlos, Brazil)emitting light at 810nm and an irradiance of 5.6W/cm2 wasused for ICG photoactivation. Light exposures were started2 minutes after the end of the second IV infusion. Light wasapplied over an interval of 90 seconds with a single 1.5mmspot per treatment. Twenty minutes after first light delivery,a second (and last) laser application using identical parame-ters was performed. A fundus contact lens of 1.5 mag-nification (Mainster widefield, Ocular Instruments, Inc.,Bellevue, USA) was used for visualization. The resultant2.25mm spot was placed directly on the center of the ALS.The single spot size used in this series was sufficient toprovide complete area coverage in all patients despiteminimal movements during the extended procedure. Retreat-ments were not performed.

Documentation

Patients with best-corrected VA 20/32 secondary to per-sistent subretinal fluid involving the macula 4 months afterinitial presentation were offered to IMP. Patients were eval-uated using a standardized regimen at 72 hours before treat-ment and 2 days, 1 and 2 weeks after treatment, as well as at

ICG-mediated photothrombosis in persistent CSC 289

3, 6, and 12 months after the IMP session. Best-correctedvisual acuity was measured using a retroilluminated Light-house for the Blind distance visual acuity test chart (usingmodified Early Treatment Diabetic Retinopathy Study charts1, 2, and R). For fluorescein and ICG angiography a funduscamera (TRC-50IA/IMAGEnet; Topcon, Tokyo, Japan) was used. For fluorescein angiography, pictures were takenbefore and at 10 seconds to 2 minutes, 2, 5, and 10 minutesafter IV administration of 5ml of a 10% sodium-fluoresceinsolution. For ICG angiography, images were taken before and at 10 seconds to 2 minutes, 5, 10 and 20 minutes afterIV administration of 25mg ICG solution (approximately 0.3mg/Kg body weight). Optical coherence tomographyevaluation (OCT 2000, Humphrey Instruments, San Leandro,CA) consisted of two linear scans 5.00mm in length centeredon the foveal region at 0 and 270 and a 7.00 scan posi-tioned along the greatest linear dimension of the serousmacular detachment. Measurement of the visual field usingHumphrey field analyzer 102 program (Zeiss Humphrey,Dublin, CA) was performed at the 3 month follow-up visit,in 3 patients in whom the laser was applied within the fovealregion.

Outcome measurements

Follow-up best-corrected VA determinations and change invision from baseline were used to evaluate the effects of theprocedure. The non-parametric Friedman test with multiplecomparisons was used for statistical analysis of the meanchange in VA since the data was not normally distributed.The significant level adopted was 0.05. Any loss in VA afterthe procedure was considered significant when evaluating thetreatment response. Data on adverse ocular events wereobtained from case records forms and inspection of fluores-cein and ICG angiography and OCT documentation.

Results

Ten out of eleven patients who were enrolled completed the1-year study. Two patients missed the 6 month follow-upvisit. The three women and eight men ranged in age from 30to 47 (mean, 38) years. Fluorescein and ICG angiography andOCT showed that at baseline, all study eyes had persistentsymptoms associated solely with CSC and that all eyes hadincreased retinal elevation caused by subretinal fluid accu-mulation. Best-corrected VA at baseline ranged from 20/100to 20/32 - 1 (mean, 20/63 + 2 [logMAR equivalent, 0.460 0.155]). Fluorescein angiography showed the characteristicfocal leak at the level of the RPE in all 11 cases on baselineexamination. All ALS identified with fluorescein angiogra-phy in these patients demonstrated increased ICG uptake onICG angiography; however, a better demarcation of the ALSwas noted in 4 cases with ICG angiography. The areas evi-denced in ICG angiography were significant larger than thoseidentified on fluorescein angiography. We found that the

active leakage area on ICG angiograms was never larger than1600 microns, regardless of the size of the macular detach-ment. In two patients the leakage area was located within thejuxtafoveal or subfoveal area; in the remaining nine eyes, theALS was located in the papillomacular bundle (n = 4) or else-where in the posterior pole (n = 5) (Table 1).

The 2.25mm (measured on the retina) laser spot used forthe IMP procedure was applied within the foveal region inthree eyes and in the macular area sparing the fovea in fiveothers; the remaining three eyes had direct treatment withinthe papillomacular bundle. At the end of the IMP sessionmild retinal whitening was observed in two eyes that hadlaser application within the papillomacular bundle.

At one week after treatment, best-corrected VA hadimproved from a mean of 20/63 + 2 at baseline to a mean of20/32 - 1, representing a mean change in VA from baselineof 2.3 (0.8) lines. Two weeks after IMP, the mean VAincrease of 3.4 (1.3) lines over baseline was statistically sig-nificant (p < 0.05). All but one patient presented visionrestoration to 20/25 levels or more 1 month after treatment.The greatest VA change from baseline (+4.3 lines) was notedat 3 months (mean best-corrected VA = 20/20 - 2). No furthersignificant changes in vision were noted during the follow-up visits 6 and 12 months after IMP. Vision change observedin the eleven eyes during the study is summarized in Table 2and Figure 1.

Fluorescein angiography indicated that all patients haddecrease in the extent of active leakage 1 week after treat-ment. At the final (12-month) follow-up visit, all patients hadsome degree of RPE mottling in the site previously identi-fied as the leakage area. Marked RPE damage was observedin two patients who presented mild retinal whitening at theend of the IMP session placed within the papillomacularbundle. Interestingly, IMP-induced damage in these patientsat the RPE level respected the limits of previously areas iden-tified on ICG baseline angiograms as the ALS in spite of the use of a laser spot quite larger than these areas (Fig. 2);associated retinal capillary obliteration with some degree of staining of the retinal vessels wall over the center of theALS was also noted in one case. We should mention that inno case the area of RPE alteration increased in size over thefollow-up period.

ICG angiography confirmed selective hypoperfusion of the leakage area. Choroidal occlusion after IMP wasrestricted to the area previously identified as the ALS in allpatients, and no detectable angiographic alterations in thenormal choriocapillaris and choroidal vessels in the vicinityof the ALS included in the treatment spot was seen (Fig. 3).

OCT findings agreed with those of fluorescein and ICGangiography in showing a gradual decrease in retinal eleva-tion due to reduced or absent of leakage after IMP in all eyesas early as 1 week. On follow-up OCTs, no detectable alter-ation was noted in the reflective bands corresponding to theretinal layers and RPE/choriocapillaris complex in 8 patients(Fig. 4). In the remaining 3 patients abnormal decrease inretinal thickness over the former ALS was observed in one


Table 1. Baseline characteristics and best-corrected visual acuity (Snellen / LogMAR).

Patient number, age, gender, (Snellen / LogMAR)

study eye, ALS, laser application Baseline Day 2 Week 1 Week 2 Month 1 Month 3 Month 6 Month 12

1, 47, F, RE, S, F 20/100 + 1 20/80 + 1 20/40 - 1 20/25 20/20 - 1 20/20 - 1 20/20 20/20 - 10.68 0.58 0.32 0.1 0.02 0.02 0 0.02

2, 30, M, RE, PB, PB 20/63 - 1 20/50 - 2 20/40 20/25 + 2 20/25 + 2 20/20 - 2 20/25 + 2 20/25 + 20.52 0.44 0.3 0.06 0.06 0.04 0.06 0.06

3, 31, F, RE, PB, F 20/63 + 2 20/40 - 2 20/32 20/20 20/20 20/20 20/20 20/200.46 0.34 0.2 0 0 0 0 0

4, 42, F, LE, PB, PB 20/40 + 1 20/32 20/20 - 2 20/20 20/20 20/20 * 20/200.28 0.2 0.04 0 0 0 * 0

5, 38, M, RE, EPP, EPP 20/63 20/63 + 2 20/40 - 1 20/25 20/25 + 2 20/20 - 2 20/20 - 2 *0.5 0.46 0.32 0.1 0.06 0.04 0.04 *

6, 33, M, LE, EPP, EPP 20/80 + 2 20/63 20/32 20/25 - 2 20/25 - 2 20/20 20/20 20/20 - 10.56 0.5 0.2 0.14 0.14 0 0 0.02

7, 32, M, RE, PB, PB 20/100 20/100 20/100 + 2 20/80 20/40 20/25 - 2 20/32 + 2 20/25 - 20.7 0.7 0.66 0.6 0.3 0.14 0.16 0.14

8, 42, M, LE, EPP, EPP 20/50 - 2 20/40 20/32 + 1 20/25 - 2 20/25 20/25 * 20/250.44 0.3 0.18 0.14 0.1 0.1 * 0.1

9, 39, M, RE, J, F 20/32 - 1 20/25 20/20 - 1 20/20 20/20 20/20 - 1 20/20 20/20 - 20.22 0.1 0.02 0 0 0.02 0 0.04

10. 42, M, RE, EPP, EPP 20/50 20/40 - 1 20/32 20/25 + 1 20/20 - 1 20/20 20/20 20/200.4 0.32 0.2 0.08 0.02 0 0 0

11, 43, M, LE, EPP, EPP 20/40 20/40 + 2 20/25 - 1 20/20 - 2 20/20 20/20 20/20 20/200.3 0.26 0.12 0.04 0 0 0 0

Mean BCVA 20/63 + 2 20/50 + 1 20/32 - 1 20/25 - 1 20/25 + 2 20/20 - 2 20/20 - 1 20/20 - 20.46 0.381818 0.232727 0.114545 0.063636 0.032727 0.028889 0.038

1 standard deviation 0.155177 0.174231 0.174189 0.169255 0.091134 0.04671 0.053955 0.048488

ALS = active leakage site location; F = female; M = male; RE = right eye; LE = left eye; BCVA = best-corrected visual acuity; S = subfoveal;J = juxtafoveal; PB = papillomacular bundle; EPP = elsewhere in posterior pole; F = foveal.* No visual acuity measurement for this interval.

Table 2. Visual acuity change (LogMAR) from baseline by visit.

Patient number Day 2 Week 1 Week 2 Month 1 Month 3 Month 6 Month 12

1 0.1 0.36 0.58 0.66 0.66 0.68 0.662 0.08 0.22 0.46 0.46 0.48 0.46 0.463 0.12 0.26 0.46 0.46 0.46 0.46 0.464 0.08 0.24 0.28 0.28 0.28 * 0.285 0.04 0.18 0.4 0.44 0.46 0.46 *6 0.06 0.36 0.42 0.42 0.56 0.56 0.547 0 0.04 0.1 0.4 0.56 0.54 0.568 0.14 0.26 0.3 0.34 0.34 * 0.349 0.12 0.2 0.22 0.22 0.2 0.22 0.18

10 0.08 0.2 0.32 0.38 0.4 0.4 0.411 0.04 0.18 0.26 0.3 0.3 0.03 0.3

Mean 0.078182 0.227273 0.345455 0.396364 0.427273 0.423333 0.4181 standard deviation 0.041429 0.088667 0.134191 0.117241 0.138643 0.193261 0.1465

* No visual acuity measurement for this interval.


eye, and increased reflectivity from choroid consistent withsome RPE atrophy in two eyes.

The measurement of the central visual field in patientswho had laser treatment over the foveal region 3 months afterIMP, demonstrated normal foveal sensibility (35, 35, and 37dB) and absence of central scotoma. Overall, the treatmentdid not cause any systemic photosensitivity complications,skin photosensitivity reactions, or major ocular adverseevents such as nonperfusion in retinal vessels.

Discussion

Laser treatment for the management of exudative manifes-tations in CSC remains a controversial issue, particularlybecause of the favorable visual outcome commonly observedafter spontaneous disease remission.24 Nevertheless, thepresence of persistent pigment epithelial detachment and/orsubretinal fluid have been suggested to be related withreduced final VA.68 Results of alternative therapeutic modal-ities have been controversial and so far no definite provenmethod is generally accepted.922 In the prospective caseseries we report, IMP led to rapid improvement in visualfunction, as documented by change in best-corrected VA,with no significant complications, in all patients with persis-tent CSC who underwent the procedure. Although only a fewpatients were included in this series, the mean change invision was statistically significant as early as 2 weeks afterthe procedure (p < 0.05). The improvement in vision wasstrongly related to decreased activity and leakage cessationfrom the former ALS as documented by fluorescein and ICG

angiography leading to restoration of the macular architec-ture as seen on OCT evaluations. As far as we are aware, thisis the first study to propose continuous 810nm laser andintravenous ICG infusion as a treatment for persistent CSC.

Of the various reports about the use of different laser treat-ments for the management of CSC,922 no definitive agree-ment existed among the real benefits induced by laserphotocoagulation. Indication for laser therapy is based on itshigh rate of success in decreasing symptoms duration, andits minimal damage induced by the treatment (usually limitedto the extrafoveal area). Although claimed quite safe andeffective, laser treatments achieved limited results in themanagement of chronic, severe, or recurrent CSC variantsthat were associated with persistent subretinal fluid andcentral RPE atrophy.13,25 In addition, complications related tolaser photocoagulation such as accidental foveal damage,retinal distortion, significant loss of contrast sensitivity, andchoroidal neovascularization, have been also reported.14,22,26

We believed we could overcome these last, and some others,limitations to laser photocoagulation of CSC just discussed.First, we analyzed fluorescein and ICG angiograms at initialand baseline evaluations and found that: (1) an increase in ICG uptake existed in ALS identified on fluoresceinangiograms; and (2) ALS were greater and easier to delin-eate in 4 out of the eleven patients in ICG angiograms.Second, we proposed a new approach to ALS treatment, toaddress the factors previously mentioned that have beenreported to have a negative impact on the success of highintensity (W/cm2) laser treatments. By the use of low-intensity, continuous application of 810-nm laser and ICG wehave already demonstrated that effective and relatively selec-

Figure 1. Visual acuity change (LogMAR) from baseline by visit.


tive occlusion of the choriocapillaris23 and neovascularlesions complicating AMD24 could be achieved. In addition,utilizing a similar diode laser, Mainster and Reichel sug-gested that irradiances of up to 11.0W/cm2 can be safelydelivered to the fundus over 60 seconds for the treatment ofoccult subfoveal CNV,27 and Connolly et al.28 describedlimited outer retinal changes or no damage at all byhistopathologic study in otherwise normal retinal tissue inhuman eyes after prolonged irradiances of 9.0W/cm2. Takentogether, these points establish the rationale of IMP for thetreatment of persistent CSC.

The fact that the laser used for IMP is the same utilizedfor hyperthermic treatments (TTT) might generate somedoubts about the mechanisms of action involved in the pro-

cedure. Studies on tumor vessel damage after laser-inducedhyperthermia demonstrated an increased vessel permeabilityand extreme edema of vascular endothelial cells.29 Becausenear-infrared wavelength is not absorbed by hemoglobin, vascular damage is likely to occur by indiscriminate andunpredictable heat transmission from the endogenous pig-mented targets of the radiation (RPE cells and choroidalmelanocytes). Consequently, the absorption center liesoutside the vessel lumen, and the vascular endothelium is damaged by heat radiating from endogenous pigmentstoward the vessel wall. When using a blood-borne exogenouspigment in IMP, we could minimize, but certainly notexclude, this heat generation (arising from endogenous pig-ments) by the use of lower intensities laser irradiations for

Figure 2. (A) Typical focal leak at the level of the retinal pigment epithelium was still seen on fluorescein angiography four months afterinitial presentation. Doted line represents the 2.25 mm spot used for laser application. (B) A well-defined irregular area of hyperfluorescencecorresponding to the active leakage site (ALS) surrounded by an area of presumed choroidal hyperpermeability was seen on indocyaninegreen angiography. (C) One week after treatment subretinal fluid resolved and extensive RPE alteration was seen on fluorescein angiography.Retinal vessels over the ALS remained perfused and with no angiographic signs of damage. (D) Indocyanine green angiography revealed anirregular hypofluorescent area corresponding to the former ALS. Note that in spite of the large rounded spot utilized for treatment no damagewas evidenced on the presumed normal choroidal tissues involved in the laser spot.


Figure 3. (A) Increased serous macular detachment caused by a well defined active leakage area (ALS) was seen on fluorescein angiogra-phy four months after initial presentation. (B) Indocyanine green angiography delineated two areas of active leakage and multiple areas ofhyperfluorescence consistent with presumed choroidal hyperpermeability. (C) One week after treatment minimal alterations at the retinalpigment epithelium level was seen and almost complete restoration of the macular architecture was observed. (D) Indocyanine green angiog-raphy demonstrated selective occlusion of the ALS previously identified. (E, F) Note that in spite of the large spot used for laser applicationover the foveal region, no corresponding scotoma was seen three months after treatment in 102 visual field measurement. Foveal sensibilityat that time was normal (35 dB).

the procedure. Therefore the absorption center lies within the region that present higher concentrations of the dye whosepeak absorption matches the laser light wavelength. By divid-ing the ICG injections in two dye-bolus infusions, twentyminutes apart, we theoretically maximized photochemicaleffects at the desired target as follows: (1) first ICG infusion

(loading dose) leads to ICG staining of the walls of thechoroidal vessels which present abnormal permeabiityleading to breakdown of the outer blood-retinal barrier as demonstrated by the results of the angiographic studiesherein and by Guyer et al.;5 (2) second ICG infusion leads toa high concentration of the dye in the choroidal intravascular


doses of light (J/cm2); in theory the desired induced effect inTTT is considered an adverse effect in IMP (Fig. 5).

The procedure sequence and laser irradiance used in thisseries (5.6W/cm2) were chosen based on preliminary resultsof the studies using IMP in exudative AMD (Vitreous SocietyMeeting 2001). Even though this level could have causedside effects such as irreversible damage to the inner retinallayers of the fundus in the targeted area, we believed that therisk of this effect was lower with IMP because the intensitywe used in this study was at least 100 times lower than thatregularly used for laser photocoagulation. Experimentalobservations and theoretical models have suggested thatintensity is the main factor contributing to temperaturechanges in laser-treated tissues, and this is borne out by the

Figure 4. Optical coherence tomography (OCT) performed 72hours before treatment and 2 days, 1 and 2 weeks, and 1 month afterindocyanine green-mediated photothrombosis (White bar = 250micras). (A) Sequential 5.00 mm horizontal OCT scans demon-strated rapid and gradual restoration of the macular architecture asevidenced by the resolution of the subretinal fluid as well asdecrease in retinal thickness. (B) 7.00 mm scans of another patientwho had treatment over the foveal region. A similar course wasobserved and no alteration at the level of the retinal pigment epithe-lium and choriocapillaris hyperreflective band was seen.

Figure 5. Schematic sequence in indocyanine green (ICG) medi-ated photothrombosis and hyperthertmic treatment. Prior to treat-ment the choroidal vessels that present abnormal permeabilityleading to fluid extravasation and retinal elevation (active leakagesite [ALS]) are represented in yellow. (A) First ICG infusion(loading dose) leads to ICG staining of the walls of the choroidalvessels which present abnormal permeability after 20 minutes(green rings). (B) Second ICG infusion 2 minutes prior to laserapplication leads to a high concentration of the dye in the choroidalintravascular compartment (green circles). (C) The use of a blood-borne exogenous pigment (ICG) minimized the heat generation (red dots) arising from endogenous pigments by the use of lowerintensities 810 nm laser irradiations. (D) The absorption center lieswithin the region that present higher concentrations of the dyewhose peak absorption matches the laser light wavelength, induc-ing a higher photochemical effect (blue) in those areas that wereformerly ICG-loaded. (C) When no dye is used in continuous810 nm laser application (TTT), vascular damage is likely to occurby indiscriminate and unpredictable heat transmission (red dots)from the endogenous pigmented targets of the radiation becausenear-infrared wavelength is not absorbed by hemoglobin; further-more a greater heat generation is expected to occur since higherlaser intensity is used in such procedure. (D) The absorption centerlies outside the vessel lumen, and the vascular endothelium isdamaged by heat radiating from endogenous pigments toward thevessel wall (blue).

compartment. Although the second ICG injection promotesrandom intravascular distribution after two minutes (whenlight delivery initiates) a higher photochemical effect occursrather in those vessels/areas involved in the treatment spotthat have been formerly ICG-loaded (the ALS in this series)than in the normal choroidal vessels that presents minimal orno ICG residual up-take from the first ICG injection. Wecould have achieved an even higher effect with the laser appli-cation initiating seconds after the second infusion because ofthe rapid ICG clearance, but a two-minute interval waschosen to allow some decrease in ICG concentration of theretinal circulation. Thus, similarities among IMP and hyper-thermia (TTT) are restricted to the fact that both use the samelaser wavelength for continuous light application and similar


relative effects of various lasers used clinically. A faster andmore intense increase in tissue temperature occurs when con-ventional photocoagulation is used than when much higherenergy (measured in J/cm2) is delivered in a continuous pulsewith lower intensity (measured in W/cm2) as used for ther-motherapy for exudative age-related macular degenerationfor example.27

Another issue to be addressed is related to the photo-chemical response of ICG under the conditions herein pro-posed. Following laser photoactivation, a dye molecule thathas absorbed laser light energy can reach the ground state by either radiative or non-radiative decay. In non-radiativedecay, the molecules absorbed energy can be converted toheat (internal conversion) and transferred to other molecules(photooxidation type I), thereby damaging cells by raisingtheir intracellular temperature, as shown by the use of ICGfor tissue welding.30 Alternatively, the dyes absorbed energycan be transferred to molecular oxygen (photooxidation typeII) via a triplet state,31 which then interacts with oxygen andother compounds to form reactive intermediates, such assinglet oxygen, which can cause irreversible destruction ofbiologic substrates.32,33 Damage induced by singlet oxygenformation is restricted to the immediate vicinity of the pho-toactivated drug because it possesses a reactive distance ofonly 0.1mm.32 Depending on specific dye properties and lasersettings for its photoactivation, different responses can beachieved. ICG-mediated photodynamic effects were alreadydemonstrated in vitro31 and in vivo;23 however, the increasein the retinal irradiance (to compensate the lower maximumhuman tolerated ICG dose) necessary to induce a clinicalobserved effect in the clinical scenario make impossible todetermine exactly how thermal or photodynamic was thegenerated response. We hypothesized over a mixed type I andII photooxidation therapy to explain the observed effects inIMP. This is in fact the reason of the change in the initial ter-minology proposed to the procedure (i-PDT). More impor-tant than clarify the exact mechanism of action is to point outthat induced effects are mainly related to the biodistributionof the dye within the irradiated area, thus minimizing unde-sirable effects in the normal tissues.

For the IMP procedure a 2.25mm spot was chosen forlaser application. Even if the spot size used for light deliv-ery was larger than the ALS in all cases, the results hereinpresented suggest that a selective effect occurred and nodeleterious effects could be evidenced in presumably normaltissues within the treatment area by clinical examination, flu-orescein and ICG angiography or optical coherence tomog-raphy in 9 out of 11 patients. In the remaining 2 patients,mild retinal whitening over the ALS was observed immedi-ately after the end of IMP session, with subsequent retinalcapillary occlusion in one. Both patients had the ALS locatedin the papillomacular bundle, and at the edge of the serousmacular detachment within an area of attached retina. Thephotochemical reactions after ICG photoactivation in addi-tion to somewhat heat generated by laser-endogenouspigment interaction induced undesirable effects mostly

caused by the close relation between the choroidal ALS andexternal retinal layers in these eyes. Although some degreeof inner blood-retinal barrier breakdown was also seen in flu-orescein angiography after treatment, no evidence of damageto the nerve fiber layer was noted. Apparently, the presenceof fluid over the ALS was essential to achieve optimal resultsas demonstrated for instance, in 3 patients who had treat-ments placed over the foveal region. IMP-induced effects ledto VA levels of 20/20 or 20/20 - 1 as early as 1 month fol-lowing treatment; furthermore, the foveal sensibility was 35dB and no evident scotoma was demonstrated in theadditional 10.2 visual field measurement performed in allthree patients 3 months after the procedure. Apparently if the retinal layers were not primarily affected during laserapplication, laser-dye-induced effects on the RPE andchoroidal tissues were not sufficient to induce clinically significant perceptible damage at these levels.

Notable VA change occurred after IMP in all patients inthis small series. A mean statistically significant recovery of3.4 lines was observed as rapid as two weeks within treat-ment, and by the end of the 1-month study VA levels of 20/25was observed in 10 out the 11 patients. No recurrences wereobserved in the one year follow-up period with all but one(loss of follow-up visit) patient retaining vision at that time.OCT findings agreed with the clinical outcome, as demon-strated by the rapid and gradual resolution of the subretinalfluid evidenced after treatment. The clinical and OCT coursefollowing IMP was quite similar among patients with theexception of one patient (n.7) who presented a delayed recov-ery of the VA caused by late resolution of the subretinal fluid.In studies reported by Klein et al.3 and Gilbert et al.,13 ofpatients with CSC approximately 33% to 50% experiencedrecurrences after the primary episode, and approximately50% of the recurrences occur within one year of the firstepisode.19 We can not compare these results to thoseobserved in our patients because of significant study dif-ferences. Nevertheless, no patient experienced new CSCepisode during the 12 months period of follow-up. Further-more, it is difficult to determine whether the recurrence isdue exacerbation of a previous leakage, an entirely newleakage point or incomplete identification of the originalASL. Taking into consideration the greater ALS observed onICG angiograms in 36% of the patients in our study, coupledwith the fact that the new leakage site is located within a dis-tance no greater than 1mm of the original ALS in 80% ofrecurrent cases,13,19 absence of recurrences in our study mightbe related to the use of large spot for laser application; the2.25mm retinal spot centered on the ALS used may have leadto unintentional treatment of small leakage areas impossibleto detect on fluorescein angiography.

In conclusion, we have demonstrated that with the use ofIMP, a simple outpatient procedure with minimal risks andrelatively low cost, significant and rapid short-term improve-ment in VA can be achieved in patients with persistent CSC.Recovery of vision was strongly related to macular subreti-nal fluid resolution in all patients. Our study has several lim-


itations, including the absence of a control group, the smallnumber of patients, and the short period of follow-up. Nev-ertheless, the results presented in this report suggest thatfurther studies are warranted to assess the clinical value ofthis technique and to provide a better understanding of themechanisms of action involving IMP in the management ofpersistent CSC.

References

1. Straatsma BR, Allen RA, Pettit TH. Central serousretinopathy. Trans Pacific Coast Oto-Ophthalmol Soc.1966;47:107125.

2. Gass JDM. Pathogenesis of disciform detachment of theneuroepithelium: II. Idiopathic central serous choroidopa-thy. Am J Ophthalmol. 1967;63:587615.

3. Klein ML, Van Buskirk EM, Friedman E, Gragoudas E,Chandra S. Experience with nontreatment of central serouschoroidopathy. Arch Ophthalmol. 1974;91:247250.

4. Castro-Correia J, Coutinho MF, Rosas V, Maia J. Long-termfollow-up of central serous retinopathy in 150 patients. DocOphthalmol. 1992;81:379386.

5. Guyer DR, Yannuzzi LA, Slakter JS, Sorenson JA, Ho A,Orlock D. Digital indocyanine green videoangiography of central serous chorioretinopathy. Arch Ophthalmol.1994;112:10571062.

6. Yannuzzi LA, Shakin JL, Fisher YL, Altomonte MA.Peripheral retinal detachments and retinal pigment epithe-lial atrophic tracts secondary to central serous pigmentepitheliopathy. Ophthalmology. 1984;91:15541572.

7. Levine R, Brucker AJ, Robinson F. Long-term follow-up ofidiopathic central serous chorioretinopathy by fluoresceinangiography. Ophthalmology. 1989;96:854859.

8. Loo RH, Scott IU, Flynn HW Jr, Gass JD, Murray TG,Lewis ML, Rosenfeld PJ, Smiddy WE. Factors associatedwith reduced visual acuity during long-term follow-up ofpatients with idiopathic central serous chorioretinopathy.Retina. 2002;22:1924.

9. Watzke RC, Burton TC, Leaverton PE. Ruby laser photo-coagulation therapy of central serous retinopathy: I. A con-trolled clinical study; II. Factors affecting prognosis. TransAm Acad Ophthalmol Otolaryngol. 1974;78:OP205OP211.

10. Watzke RC, Burton TC, Woolson RF. Direct and indirectlaser photocoagulation of central serous choroidopathy. Am J Ophthalmol. 1979;88:914918.

11. Leaver P, Williams C. Argon laser photocoagulation treat-ment in central serous retinopathy. Br J Ophthalmol.1979;63:674677.

12. Robertson DM, Ilstrup D. Direct, indirect, and sham laser photocoagulation in the management of central serous chorioretinopathy. Am J Ophthalmol. 1983;95:457466.

13. Gilbert CM, Owens SL, Smith PD, Fine SL. Long-termfollow-up of central serous chorioretinopathy. Br J Oph-thalmol. 1984;68:815820.

14. Robertson DM. Argon laser photocoagulation treatment in central serous chorioretinopathy. Ophthalmology.1986;93:972974.

15. Slusher MM. Krypton red laser photocoagulation inselected cases of central serous chorioretinopathy. Retina.1986;6:8184.

16. Ficker L, Vafidis G, While A, Leaver P. Long-term follow-up of a prospective trial of argon laser photocoagulation inthe treatment of central serous retinopathy. Br J Ophthal-mol. 1988;72:829834.

17. Piccolino FC. Laser treatment of eccentric leaks in centralserous chorioretinopathy resulting in disappearance ofuntreated juxtafoveal leaks. Retina. 1992;12:96102.

18. Burumcek E, Mudun A, Karacorlu S, Arsln MO. Laser photocoagulation for persistent central serous retinopathy.Ophthalmology. 1997;104:616622.

19. Spitznas M. Central serous retinopathy. In: Ryan SJ, ed.Retina. St. Louis: CV Mosby; 1989, vol. 2, pp. 217227.

20. Gass JDM. Symposium: Current status of photocoagulationof macular disease. Photocoagulation treatment of idio-pathic central serous choroidopathy. Trans Am Acad Oph-thalmol Otolaryngol. 1977;83:OP456OP467.

21. Schatz H, Yannuzzi LA, Gitter KA. Subretinal neovascu-larization following argon laser photocoagulation treatmentfor central serous chorioretinopathy: Complication or misdiagnosis? Trans Am Acad Ophthalmol Otolaryngol.1977;83:OP893OP906.

22. Khosla PK, Rana SS, Tewari HK, Azad RU, Talwar D. Eval-uation of visual function following argon laser photocoag-ulation in central serous retinopathy. Ophthalmic SurgLasers. 1997;28:693697.

23. Costa RA, Farah ME, Freymuller E, Morales PH, Smith R,Cardillo JA. Choriocapillaris photodynamic therapy usingindocyanine green. Am J Ophthalmol. 2001;132:557565.

24. Costa RA, Farah ME, Cardillo JA, Belfort Jr. R. Photody-namic therapy with indocyanine green for occult subfovealchoroidal neovascularization caused by age related maculardegeneration. Curr Eye Res. 2001;23:271275.

25. Brancato R, Scialdone A, Pece A, Coscas G, Binaghi M.Eight-year follow-up of central serous chorioretinopathywith and without laser treatment. Graefes Arch Clin ExpOphthalmol. 1987;225:166168.

26. Gass JDM. Stereoscopic atlas of macular disease. 4th ed.St Louis: Mosby, 1997:5270.

27. Mainster MA, Reichel E. Transpupillary thermotherapy forage-related macular degeneration: Long-pulse photocoagu-lation, apoptosis, and heat shock proteins. Ophthalmic SurgLasers. 2000;31:359373.

28. Connolly B, Regillo C, Eagle R, Shields C, Shields J,Moran H. The histopathologic effects of transpupillary thermotherapy in human eyes. Invest Ophthalmol Vis Sci.2001;42(suppl):513. ARVO abstract.

29. Liu DL, Andersson-Engels S, Sturesson C, Svanberg K,Hakansson CH, Svanberg S. Tumour vessel damage resulting from laser-induced hyperthermia alone and incombination with photodynamic therapy. Cancer Lett.1997;111:157165.


30. Decoste S, Farinelli W, Flotte T, Anderson R. Dye-enhancedlaser welding for skin closure. Laser Surg Med. 1992;12:2532.

31. Baumler W, Abels C, Karrer S, Wei T, Messmann H,Landthaler M, Szeimies R.-M. Photo-oxidative killing ofhuman colonic cancer cells using indocyanine green andinfrared light. Br J Cancer. 1999;80:360363.

32. Henderson BW, Dougherty TJ. How does photody-namic therapy work? Photochem Photobiol. 1992;55:145157.

33. Roberts W, Hasan T. Role of neovasculature and vascularpermeability on the tumor retention of photodynamicagents. Cancer Res. 1992;52:924930.

Documents

Indocyanine TT of Chronic CRSC