Katharina Krepler MD Medical University of Vienna Department of Ophthalmology Waehringer Guertel 18–20 A-1090 Vienna Austria Tel: + 43 1 40 400/7931 Fax: + 43 1 40 400/7932 Email: firstname.lastname@example.org
Purpose: To evaluate treatment of macular oedema due to central retinal vein occlusion (CRVO) with intravitreal triamcinolone acetonide.
Methods: In a prospective case series, 13 patients with macular oedema due to non-ischaemic CRVO received an intravitreal injection of 4 mg triamcinolone acetonide. Examination included assessment of best corrected visual acuity (BCVA) for distance and reading, measurement of intraocular pressure (IOP), fluorescein angiography and high resolution imaging by optical coherence tomography, preoperatively and 1 week, 1 month, 3, 6 and 9 months postoperatively.
Results: Preoperative mean BCVA was 0.9 ± 0.4 for distance vision and 1.0 ± 0.3 for reading acuity, respectively. A significant improvement in distance VA (0.5 ± 0.3, p < 0.001) and reading acuity (0.7 ± 0.3, p = 0.03) was observed until 3 months and 6 months, respectively. Mean macular thickness was significantly reduced until 9 months postoperatively.
Conclusion: Intravitreal injection of triamcinolone acetonide led to a significant improvement in mean VA in patients with macular oedema due to CRVO. However, the significant effect was not permanent and persisted for a maximum of 6 months.
Venous occlusive disease is commonly seen in clinical practice and is a cause of significant visual morbidity. A potential complication of central retinal vein occlusion (CRVO) is macular oedema (ME). It results from leakage of the perifoveolar small vessels secondary to hydrostatic stress and ischaemia, and may occur in mild or severe cases of CRVO. Persistent ME due to CRVO has a poor prognosis (Glacet-Bernard et al. 1996) and no routine treatment has been established so far. In the Central Vein Occlusion Study Group (1995), grid laser photocoagulation had no significant impact on final visual acuity (VA).
Because there is no proven treatment of ME in CRVO other treatment options are needed. Corticosteroids have long been used for their antiphlogistic and antiproliferative effect. Triamcinolone acetonide (TA), a long-acting corticoid derivate, has been used in ocular pathologies administered either locally or systemically. It has an antiangiogenetic and anti-inflammatory effect, reduces the permeability of endothelial cells and stabilizes the blood–retinal barrier (Jonas et al. 2004).
The purpose of the present study was to analyse the clinical course and VA of patients with ME due to CRVO after intravitreal injection of TA.
This study was designed as a prospective interventional case series. It was carried out in keeping with local legal requirements and in accordance with the guidelines of the Declaration of Helsinki, including current revisions. The nature of the study was explained and all patients gave written consent to participate.
Patients with ME due to non-ischaemic CRVO, and with best corrected visual acuity (BCVA) of logMAR 0.4 or worse were chosen for inclusion in the study. Exclusion criteria were history of glaucoma or steroid-related intraocular pressure (IOP) rise.
Four mg (0.1 ml) of crystalline TA (Volon A®; Bristol-Myers Squibb SpA, Anagni, Italy) – without prior removal of the vehicle from the TA suspension − were injected into the vitreous cavity using a sharp 27-gauge needle through the inferior pars plana (4 mm from the limbus in phakic eyes and 3.5 mm in pseudophakic eyes). For local anaesthesia lidocaine 4% eyedrops were applied, followed by topical application of povidone-iodine 2.5% eyedrops (Betaisodona®; Mundipharma Gmbh, Limburg/Lahn, Germany). After surgery eyes were patched with dexamethasone and gentamicin ointment (Dexagenta POS®; Ursapharm GmbH, Saarbrücken, Germany) overnight. Postoperatively dexamethasone and gentamicin eye drops (Dexagenta POS®; Ursapharm GmbH, Saarbrücken, Germany) were applied three times daily for 1 week.
Examinations were performed at baseline and at 1 week, 1, 3, 6 and 9 months after surgery. These included assessment of BCVA, slit-lamp examination and ophthalmoscopy, measurement of IOP using Goldmann applanation tonometry, optical coherence tomography (OCT) and fluorescein angiography. For distance VA, BCVA in logMAR was assessed using ETDRS charts. Reading acuity was assessed using German-language logarithmic Radner Reading Charts®. With this test, reading acuity is defined as logRAD (logarithm of reading acuity determination), representing a reading acuity score (Radner et al. 1998). At each follow-up, macular thickness was measured using OCT II Version 2010 (Zeiss Humphrey, New York, USA) after maximal pupil dilation. The macular thickness was automatically calculated by computer software using the aggregate data from multiple radial scans (e.g. using a macular thickness map function). During the assessment patients were asked to fixate on the central green fixation target.
Main outcome measures were BCVA for distance and reading, and macular thickness assessed with OCT.
After a study period of at least 3 months, if there was clinical evidence of ME on fundus examination, a significant increase in macular thickness assessed by OCT, and VA had decreased by two or more lines after an initial improvement in VA of two or more lines, patients were offered another course of treatment with intravitreal triamcinolone. Results of the follow-up examinations after the second treatment course in these patients were not included in this study. Patients who had other ocular surgery during follow-up were also excluded.
Statistical calculations were done using standard software (MS Excel 2002, Microsoft Co.; and Statistica 6.0; StatSoft Inc, Tulsa, USA). Data were compared using paired t-test. After Bonferroni−Holm correction a p-value ≤ 0.05 was considered significant.
Thirteen patients with ME due to CRVO were included in the study between March 2002 and July 2003. The mean age of the patients was 64.9 ± 12.5 years. The duration from diagnosis of retinal vein occlusion with ME to treatment with triamcinolone was 6.0 ± 5.2 months (mean ± SD; range 3 weeks−18 months). All patients had non-ischaemic CRVO at baseline. Two patients developed ischaemic CRVO during follow-up, one with neovascularization of the disc, and one with neovascularization of the iris and secondary glaucoma. Both patients were treated with panretinal laser photocoagulation, the second patient additionally with transscleral diode laser cycloablation.
All 13 patients were completely followed up until 6 months. Two patients with recurrent ME and a decrease in VA 4 months after treatment received a second injection of triamcinolone, which led to a similar visual improvement as the first treatment course. One patient was excluded from the study 8 months after treatment because he developed an epiretinal membrane and underwent pars plana vitrectomy and membrane peeling. One patient was lost to follow-up 7 months postoperatively.
The mean BCVA for distance vision was 0.9 ± 0.4 preoperatively, and 0.6 ± 0.3 (p = 0.006) at 1 week, 0.6 ± 0.3 (p = 0.014) at 1 month, 0.5 ± 0.3 (p = 0.0007) at 3 months, 0.7 ± 0.3 (p = 0.22) at 6 months and 1.0 ± 0.5 (p = 0.7) at 9 months postoperatively. Mean BCVA for reading vision was 1.0 ± 0.3 preoperatively, and 0.8 ± 0.2 (p = 0.032) at 1 week, 0.6 ± 0.3 (p = 0.005) at 1 month, 0.6 ± 0.3 (p = 0.003) at 3 months, 0.7 ± 0.3 (p = 0.032) at 6 months and 0.8 ± 0.4 (p = 0.2) at 9 months postoperatively (Fig. 1). The mean macular thickness was 468 ± 128 µm preoperatively, and 325 ± 70 µm (p = 0.0003) at 1 week, 310 ± 52 µm (p = 0.006) at 1 month, 311 ± 123 µm (p = 0.009) at 3 months, 365 ± 157 µm (p = 0.44) at 6 months, and 394 ± 137 µm (p = 0.01) at 9 months postoperatively (Fig. 2).
Six eyes developed IOP values of 22 mmHg or more, three of them of ≥ 30 mmHg during follow-up. One of those eyes had secondary glaucoma due to neovascularization of the iris. In five eyes (38.5%) elevation of IOP occurred between 1 and 4 weeks after treatment, and was assumed to be a side-effect of triamcinolone. Five eyes required topical IOP-lowering medication, four of them until the last follow-up (including one eye with neovascular glaucoma). No eye required cataract surgery during follow-up. Minimal, mostly subcapsular, cataract occurred in four eyes, pre-existing cataract progressed in one eye, and two eyes were primarily pseudophakic. No serious side-effects, such as postoperative endophthalmitis or retinal detachment, occurred.
This study shows that intravitreal TA injection is effective in reducing ME due to CRVO and leads to a significant visual improvement up to 6 months after surgery. Intravitreal TA injection has been studied in various ocular pathologies including diabetic ME (Martidis et al. 2002; Jonas et al. 2003a; Massin et al. 2004) and ME due to retinal vein occlusion (Greenberg et al. 2002; Ip & Kumar 2002; Jonas et al. 2002, 2005; Degenring et al. 2003; Ip et al. 2003; Park et al. 2003; Chen et al. 2004). A reduction of ME could be observed. Jonas et al. (2002) injected 25 mg of triamcinolone into both eyes of a patient with ME due to CRVO and found an improvement in VA in both eyes during the follow-up period of 4 months and 6 weeks, while fluorescein leakage decreased and there were no significant complications. The degree of cataract slightly increased, whereas there was no significant elevation of IOP (Jonas et al. 2002). Greenberg et al. (2002) examined the effect of intravitreal injection of 4 mg triamcinolone in a patient with bilateral ME due to CRVO and found no improvement in VA in the eye with chronic CRVO, but a significant improvement in the eye with acute CRVO. After 6 months VA declined and a second injection was performed. Optical coherence tomography showed a decrease of ME with restoration of normal macular anatomy in both eyes. There was no elevation of IOP (Greenberg et al. 2002). In another study, Ip & Kumar (2002) described the clinical course of two patients with ME from CRVO after injection of 4 mg triamcinolone. They observed an improvement in VA in both cases. In one case, concerning a patient with non-ischaemic CRVO, the improvement lasted until the 6-month follow-up visit. In the other case, involving ischaemic CRVO, the effect did not last and VA after 3 months was worse than before surgery. In a retrospective study of eight patients with CRVO, Ip et al. (2003) observed a significant anatomic improvement with OCT in the majority of patients, accompanied by a gain in VA in half the patients 3 months after treatment. Park et al. (2003) evaluated 10 eyes with cystoid ME after non-ischaemic CRVO in a retrospective study. All eyes demonstrated biomicroscopic improvement in ME, with corresponding improvement in volumetric OCT and VA, that lasted through a mean follow-up period of 4.8 months (Park et al. 2003).
In our prospective study, using a longer follow-up period, we observed a significant improvement in VA (distance vision and reading acuity) in patients with ME due to CRVO. In our opinion, assessment of distance as well as reading acuity is important if patients with macular pathology are studied. Reading is an important parameter for patients' everyday life and, therefore, improvement of reading acuity is an important therapeutic goal.
The significant effect of a single injection persisted until 6 months postoperatively but was no longer visible at 9 months. Two patients who initially reacted well to the treatment received a second injection of TA because of recurrent ME and decrease of vision 4 months postoperatively. In both of these patients, the second injection led to a similar visual improvement as the first treatment course. This observation is consistent with a case reported by Greenberg et al. (2002), whose patient received a second injection after 6 months.
The overall deterioration of VA in our patients after 6 months is possibly caused by increasing thickness of the macula after absorption of triamcinolone from the vitreous body. However, we can not rule out the possibility that cataract is also partially responsible for longterm visual deterioration (although only minor changes were observed in most patients).
The most common side-effect reported after intravitreal injection of triamcinolone is an IOP rise. In our study, six patients developed IOP ≥ 22 mmHg and three of these patients developed IOP ≥ 30 mmHg. One of these patients suffered from neovascular glaucoma. Five patients (38.5%) needed topical antiglaucomatous medication, four of them until the final follow-up visit (including one patient with neovascular glaucoma). This percentage is similar to that reported in most other studies, which found an IOP rise in 25 − 36% of patients (Challa et al. 1998; Danis et al. 2000). Jonas et al. (2003b) observed an IOP rise in a higher percentage of patients (52%) after injection of a dose of 25 mg triamcinolone.
The occurrence of posterior subcapsular cataract is a well known complication of the longterm use of topical or systemic steroids. Development or progression of cataract has been described in patients after intravitreal injection of TA. Progression of cataract was observed in six of 26 eyes in a study by Challa et al. (1998) and in four of seven eyes in a study by Danis et al. (2000). Jonas et al. (2003a) observed a progression, although not a significant one, in their patients. Park et al. (2003) observed a progression in one of 10 eyes and Martidis et al. (2002) noted similar in one of eight eyes at 6 months. However, cataract grading was not performed according to a standardized grading system and the follow-up periods were rather short in most of these studies. In our study, no patient required cataract surgery during follow-up, although beginning (mostly subcapsular) cataracts could be observed in five of 11 phakic patients (45.4%). However, we did not carry out cataract grading according to a standardized grading system. Cataract development and progression should be further evaluated in future studies because it might be an important side-effect, especially in younger patients.
Intraocular infection is a potential risk of every intraocular injection and has been observed after triamcinolone injection (Moshfegi et al. 2003; Nelson et al. 2003). There have also been reports of non-infectious endophthalmitis (Nelson et al. 2003; Roth et al. 2003; Sutter & Gillies 2003). We did not observe any case of infectious or non-infectious endophthalmitis in our patients.
Retinal toxicity of TA has been tested in animal models. Experimental studies have shown that intravitreal triamcinolone has no effect on the electroretinogram or the histology in the rabbit retina, and that it is safe in monkeys (McCuen et al. 1981; Ishibashi et al. 1985). The vehicle of triamcinolone has been shown to be devoid of any toxic effects in rabbit eyes (Hida et al. 1986). However, it has been speculated that a toxic effect of the vehicle might be responsible for the occurrence of non-infectious endophthalmitis (Roth et al. 2003). In our patients, we did not observe any clinically visible toxic or immunogenic reactions, although we did not eliminate the vehicle from the triamcinolone suspension. Recently, Nishimura et al. (2003) described a method by which almost all the vehicle and additives can be removed from the triamcinolone suspension in a simple and quick way.
Major limitations of our study include the lack of a control group and the small number of cases. However, it is the first prospective study of a series of patients with CRVO using a longer follow-up period of 9 months, and includes assessment of VA and retinal thickness using OCT.
We observed a positive effect on VA and macular thickness after intravitreal injection of 4 mg TA in patients with ME due to CRVO, but this effect persisted for 6 months only. Results of previous studies of intravitreal TA in various ocular pathologies provide no evidence that TA has a dose-related effect, but further studies are necessary to define the optimal dose of TA to be injected.
The prognosis of patients with ME due to CRVO is poor and no proven treatment exists. Therefore, intravitreal triamcinolone may be a therapeutic option, although it should be emphasized that the treatment effect of a single injection is not permanent. The potential benefit of this treatment has to be weighed up against the potential side-effects.
The paper was presented at the Association for Research in Vision and Ophthalmology (ARVO), Fort Lauderdale, Florida, USA, April 28, 2004.