Diabetic macular edema is the major cause of visual loss in patients with diabetes mellitus. (2) Up to 78% of non-insulin dependent diabetics with retinopathy have been found to have diffuse or focal macular edema. (17) Approximately one-fifth of the newly diagnosed diabetic patients develop maculopathy (36) and half of them lose one or more lines of visual acuity after follow up for 2 years. (10) Although the use of laser photocoagulation result in marked reduction in the incidence of blindness from diabetes during the past 20 years (8), in contrast, many studies reported poor prognosis after laser photocoagulation. (16) After laser photocoagulation, improvement was reported in 29.6%, while stabilization and deterioration was reported in 37% and 33.3%, respectively. (5) Early treatment diabetic retinopathy study research group (ETDRS) reported that about 24% of immediately treated eyes had thickening involving the center of the macula at 36 months (27), which encourages the interest for other treatment modalities. (35) In the last 2 years, utilization of intravitreal triamcinolone acetonide has been exponentially increased as a treatment option for various intraocular neovascular and proliferative edematous disorders. (15) Recently the risk of further deterioration of visual acuity from diffuse diabetic macular edema has been found to be reduced with repeated intravitreal corticosteroid treatment. (20) The best response was obtained in cases of intraretinal edematous diseases such as diffuse diabetic macular edema. In addition, this response is dose dependent following single intravitreal injection. (15)
Aim of the work
Our objectives in this study were: 1) to prospectively (at baseline and after follow up period of 1-, 3- and 6-months) investigate the efficacy and safety of one intravitreal injection of 4 mg (0.1 ml) of triamcinolone acetonide (IVTA), as a primary treatment in patients suffering from diffuse clinically significant macular edema (CSME) caused by diabetes mellitus, and 2) to monitor objectively the functional and anatomic improvement in patient’s vision. Measuring visual acuity (VA) and visual evoked potential (VEP) were utilized to assess and monitor visual function and Optical Coherence tomographic (OCT) was utilized to quantitatively measure macular thickness and volume.
PATIENTS AND METHODS
The study design is clinical, interventional, nonrandomized and comparative. In this study, no placebo controls were undertaken for fear of exposing subjects to risks of intravitreal injection without any benefit. Included in this study were 24 patients (24 eyes evaluated) with diffuse clinically significant macular edema (CSME). Their ages ranged from 38 to 65 years (mean; 55.25±14.94, male/female; 10/14), duration of diabetes mellitus (D.M.) ranged from 3 to 15 years (mean, 8.92±5.81) and duration of visual impairment ranged from 8 to 36 months (21.75±13.82). Patients were recruited from the outpatient Ophthalmology and Neurology clinics, Assiut University Hospital, Assiut, Egypt. This study was approved by the regional ethical committee. Informed written consent was obtained from all subjects before their participation in this study. Excluded from this study were patients with diminished visual acuity due to causes other than D.M., including, a) opaque ocular media, b) patients previously treated for their macular edema by other modalities of therapy, and c) patients with known history of high intraocular pressure (IOP) or glaucoma. All patients underwent full medical, neurological and opthalmological history and examination. Forty healthy control subjects were included for comparison. Functional and anatomic monitoring was done for all patients. The functional response was assessed by VA and VEP measurements. The anatomical response was assessed by OCT utilized to quantitatively measure the macular thickness and volume. Baseline visual acuity (VA), visual evoked potential (VEP) and Optical Coherence tomographic (OCT) assessments were undertaken pre- and post- treatment, after 1-, 3- and 6-month(s) follow up injection.
Full opthalmological examination included:
1) visual acuity measurement by Landolt broken ring chart, 2) intra-ocular pressure (IOP) measurement by Goldmann applanation tonometry. The angle of the anterior chamber was measured by Goldmann 3-mirror lens, 3) determination of lens clarity by slit-lamp examination with a +78 or +90-diopter lens, and 4) fundus examination using direct opthalmoscope and Volk +90 lens (Heidelberg Retina Angiograph, Heidelberg Engineering GmbH, Heidelberg, Germany), 5) evaluation of the macula by slit lamp biomicroscopic fundus examination for detection and assessment of the degree and grading of CSME. The latter is defined if one or more of the following characteristics was present: (a) thickening of the retina at or within 500micron of the macular center, (b) hard exudates at or within 500micron of the center of macula, if associated with thickening of adjacent retina, and (c) zone(s) of retinal thickening one disc area or larger or at any part of which is within one disc diameter of the center of macula. (35)
Visual evoked potential:
Pattern reversal visual evoked potentials (PRVEPs) with checks of 16’ of Dantec Keypoint equipment Medtronic, Copenhagen, Denmark, was done in the Neurology department, as described before. (33) VEP is a sensitive method to detect early abnormalities within the optic pathway. Recording over the mid-occiput, PRVEP usually has a negative-positive-negative configuration, in normal subjects the major positive peak occurs at 100msec (P100). Peaks are labeled using the average latency values in normal subjects: N75, P100 and N145. VEP amplitudes are more variable. We measured the amplitude as the sum of the peaks from N75 to N100 and that of P100 to N145.
Optical Coherence Tomography (OCT):
Optical Coherence Tomography (Stratus OCT 3, software version 4, Zeiss-Humphrey, Germany) provides enhanced visualization of the geometry and distribution of macular edema. It was utilized to quantitatively measure the central macular thickness and macular volume, using 850 nm superluminescent diode at maximum intensity of 750W. All eyes were scanned with 6 radial lines centered on the central fovea, with each line 5.0mm long. The location of the central fovea was determined by patient’s fixation. The macula was divided into 3 regions, including a central disc (1000m in radius), middle (3000m) and an outer ring (6000m). Each region corresponds to the fovea, parafoveal and perifoveal area, respectively. Each region was further divided into superior, inferior, temporal and nasal quadrants. The retinal thickness, defined as the length between inner retinal surface and retinal pigmented epithelium, was measured by the computer from the tomograms. The images of OCT were displayed in a false-color scale where bright colors, such as red and white indicate highly reflective areas and dark colors, while blue and blacks correspond to low reflective areas.
Patients rested in a supine position to allow the steroid to settle posterior. 5% povidone iodine solution was applied to the eye prior injection. A lid speculum was placed onto the eye and 0.5ml of 2% lidocaine was injected in the inferior temporal subconjunctival space. An individual unopened 1 ml (40mg) vial of triamcinolone acetonide (Kenalog 40; Apothecon, Princeton, NJ) was mixed and 0.1 ml (4 mg) was drawn into a tuberculin syringe. Injection was done slowly through the inferior pars plana (3.5-4mm from limbus measured with calipers) using a 27 gauge needle (Tip Box 3). The drug is injected slowly and directed posteriorly. Indirect ophthalmoscopy was used to confirm proper intravitreal localization of the suspension and perfusion of the optic nerve head. Intraocular pressure (IOP) was measured and if increased, aspiration of equal amount (0.1 ml) was undertaken through paracentesis incision down and out. The potential complications related to the injection procedure and corticosteroid medication were monitored, including IOP response, cataract progression, retinal detachment, vitreous hemorrhage, and endophthalmitis.
All the data were analyzed by the SPSS, version 10. Calculation of the normal limits of electrophysiological data was done utilizing parametric (t-test and Pearson’s correlation) and non-parammetric statistics (Spearman rank), when the distribution in normal individuals is non-gaussian (e.g. amplitude distribution). Data were set in mean±SD. Chi-square test was applied for binomial data. The differences between patients and controls and pre- and post- intravitreal injection concerning continuous variables were analyzed by Student’s t-test for unpaired samples or the Mann-Whitney U-test, when appropriate. For all tests, P<0.05 was set as significant.
In this prospective, non-comparative study, 24 patients (24 eyes evaluated) with diffuse clinically significant macular edema (CSME) were included. Demographic data are summarized in Table 1. The amplitude of visual evoked potential (VEP) was significantly lower in eyes with macular edema than in control eyes (7.31±2.23 vs.18.1±2.5 µV for control; P<0.001). But no significant difference in latency was found (99.93±2.5 vs. 98.0±2.1 msec for control) (Tables 2, 4). Best corrected visual acuity was not ascertained which makes comparisons less meaningful and do not allow determination of moderate visual loss or gain, however, we tested all eyes with the same correction throughout the follow up period. The corrected visual acuity and the electrical response density from the macular area were significantly reduced in eyes with diabetic macular edema (Table 3). Compared to controls, the mean initial pretreatment central macular thickness and the total macular volume were significantly increased (375±35.50 micron; p<0.0001 and 10.39±4.87mm3; p<0.001 respectively) (Table 3, 4). The mean baseline intraocular pressure (IOP) was 13.92±3.85 mmHg (Tables 2, 4). The pre-operative central macular thickness and the visual acuity were positively correlated with total macular volume (r=0.676, p<0.001 and r=0.715, p<0.001 respectively) and duration of visual deterioration (r=0.512, p=0.003). The pre-operative corrected visual acuity and VEP amplitudes were inversely correlated (r=0.680, p<0.001). VEP was inversely correlated with macular thickness (r=0.678, p<0.001) and visual acuity (r=0.596, p<0.001). Authors observed a significant improvement in most of the patients, both functionally and anatomically during spanning the first, third and six month(s) follow up period after treatment (Tables 2, 3, 4). Visual evoked response showed marked improvement in amplitude from initial 7.31±2.23 to 15.37±3.88 (P<0.001), 26.69±3.72 (P<0.001) and 25.65±2.28 µV (P<0.001) (Table 4). Macular thickness was reduced from initial 375±35.50 to 233.33±40.17 (P<0.001), 145.83±27.58 (P<0.001) and 202.46±29.60 micron (P<0.01). Macular volume was reduced from initial 10.39±1.87 to 7.73±1.01 (P<0.001), 6.55±0.99 (P<0.01) and 6.61±1.09 micron (P<0.01). IOP was elevated from 13.92±3.85 to 20.58±8.42 mmHg (P<0.001) at the end of the first month and decreased significantly to 15.83±4.15 mmHg and 14.63±1.46 (P<0.001, P<0.001 and P=NS) at the end of the third and sixth months follow up periods respectively (Table 4). Six patients (25%) showed decline of visual acuity at the end of 6-months follow-up periods due to recurrence of edema and necessitated reinjection (Table 3).
Macular edema is one of the important causes of visual loss in patients with diabetic retinopathy. Breakdown of the blood-retinal barrier with increased vascular permeability and fluid accumulation in the inner layer of the retinal is accepted as possible pathogenesis of diffuse diabetic macular edema. (6) Histopathological studies in macular edema indicate that fluid leaks out of the damaged retinal vessels, enters Müller cells and causes intracellular swelling, especially in the outer plexiform layer. (39) The causes of such breakdown in diabetes mellitus (D.M.) is not fully clear, however, the increased permeability and angiogenesis that occur in diabetic retinopathy was suggested to result from changes in ocular growth factors including: insulin-like growth factor-1 and its binding proteins, platelet derived growth factor, fibroblast growth factor and vascular endothelial growth factor. (1, 3)
Diabetic clinically significant macular edema (CSME) was reported to have poor prognosis with laser photocoagulation compared to focal macular edema. (20, 22) Moderate visual loss was mainly studied by the Early treatment diabetic retinopathy study research group (ETDRS) and improvement in 3 lines was observed only in <3% of patients after laser photocoagulation. (27) Recently pars plana vitrectomy has been advocated for treatment of diffuse macular edema but found to be applicable to a selected subset of patients and requires significant surgical intervention with inherent risks, recovery time and expenses. (14, 26) Application of local corticosteroids for treatment of diffuse diabetic macular edema has been shown to be effective at both experimental and human studies. (37) Corticosteroids are traditionally used for inflammatory disorders because of their ability to diminish neutrophil transmigration, limit access to sites of inflammation, and decrease cytokine production. It inhibits the arachidonic acid pathway, a product of prostaglandins. More recently, investigators have focused on the angiostatic and antipermeability properties of corticosteroids for posterior segment diseases, diabetic retinopathy, and macular edema (7) and found that in proliferative diabetic retinopathy, steroids may directly downregulates the production of growth factors such as vascular endothelial derived growth factor and inhibit leukocytes that play an important role in early microvascular alterations. Corticosteroids has been found to reduce the breakdown and stabilizes the blood-retinal barrier and hence the rationale for its introduction for treatment of diabetic macular edema. (37) Recently, corticosteroid has been found to increase barrier properties of retinal endothelial cells and at the same time induces specific changes to the tight junction proteins. It has been found to induce assembly of the synthesized tight junction proteins hence reducing the endothelial cell transport of water and solute, provides increased barrier properties and forms molecular basis that account for the change in transport properties. (29) Tight junctions create a selective barrier to water and solutes, blocking paracellular flux across the tissue and preventing lipids and proteins from diffusing between the apical and basolateral plasma membranes. The tight junctions or zonula occludens has been shown by electron microscopy as a series of anatomizing fibrils that encircle most of the apical regions of the CNS and retinal capillaries cells. (9, 34) A number of proteins have been located at tight junctions and are believed to provide the structural basis of tight junctions, as well as a range of proteins has been found to be potentially involved in the regulation of barrier permeability. (11, 23) The above results provide novel insights into the molecular basis by which corticosteroid treatment prevents tissue edema and provides a rationale for localized corticosteroid treatment for diabetic retinopathy.
Intravitreal triamcinolone (IVTA) is an emerging interventional promising therapeutic method for treatment of diabetic macular edema. (20) After intravitreal injection, the drug is delivered rapidly to its site of action with maximal bioavailability with half life of 1.6 days (32) and long lasting effect of 21 to 41 days. (30, 32) Our results highly suggest and recommend the use of IVTA as a primary treatment of diffuse diabetic macular edema. In the light of the characteristics of the studied group of patients, the results of this study indicate progressive marked observed improvement in vision when functional and anatomic assessment was done utilizing VA and VEP for visual function and OCT for quantitative anatomic evaluations. The foveal thickness measured by OCT and VEP amplitude were correlated well with the visual acuity as previously reported. (25, 31) It has been found that traditional methods of evaluating macular thickness, such as ophthalmoscopy or stereoscopic biomicroscopy, are insensitive to detect small changes in retinal thickness and intraretinal structures (40, 41), however, the advent of OCT offers the possibility of both high-resolution cross-sectional images of the retina and quantitative measurements of the retinal thickness. (12, 13) Due to the higher density of Müller cells at the foveal floor than at the retinal edges, the low reflectivity space in the outer retina of the optical coherence tomogram may represent the swollen Müller cells in the outer plexiform layer as marked retinal thickening has shown to be reflected by the abundant swollen or ruptured Müller cells in the fovea and parafoveal areas. (25, 28)
The results of this study demonstrated that the reduced amplitude of the VEP depends on the reduced visual acuity with no significant change in latency. VEP has been shown to be appropriate for objective assessment of the function of the underlying macula and evaluating the degree and progression of the disease. (4, 19, 24) Combination of OCT and VEP might be thought to provide objective criteria for the evaluation, assessment and follow up of diabetic macular edema.
Many previous studies demonstrated the safety and efficacy of IVTA in treatment of diabetic macular edema that fails to respond to conventional photocoagulation. In the study of Martidis et al. (20), IVTA was utilized to treat 16 eyes with CSDME, which was subjected to initial laser photocoagulation therapy at least 6 months before steroid injection. Authors reported improvement in mean visual acuity and macular thickness spanning 1-, 3-and 6-month(s) follow up intervals. No significant complications were observed that prohibited further injection and 3 eyes necessitated re-injection due to recurrence of edema detected after 6 months follow up period. In our study, only elevated IOP was observed during follow-up period but decline to values near baseline at the end of the third and sixth months. No other complications were reported that prohibited further injection. Six patients (25%) reported deterioration at the end of 6 months follow up due to recurrence of edema that necessitated re-injection and this is consistent with other studies. (20) The safety of intravitreal corticosteroids has been supported by many animal (21) and human studies. (18, 20, 38) The major previously reported ocular side effects include secondary ocular hypertension or glaucoma (reported in up to 40% of eyes injected) and cataract. Others include; postoperative infectious and non-infectious endophthalmitis, and pseudo-endophthalmitis. (15) Cataract progressive was usually seen with long up follow up period (>= 6 months). (20)
To our knowledge, this is the first report of utilizing intravitreal triamcinolone acetonide as a primary treatment in diffuse diabetic macular edema. Based on the above results, we concluded that 1) Intravitreal triamcinolone acetonide is safe and effective as primary treatment for diffuse chronic diabetic macular edema, however, large numbered controlled trials are needed, and 2) VEP and OCT are valuable objective diagnostic and practically simple techniques for assessment and monitoring the anatomic and functional improvement following intravitreal corticosteroid injection for treatment of diffuse diabetic macular edema.
Table 1: Pretreatment clinical characteristics
|Pat.#||Sex||age||Type of diabetes||Eye||Retinopathy||Duration of diabetes mellitus (years)||Duration of macular edema (month)|
|1||F||55||Non insulin dependent||Right||Severe NPDR||3||24|
|2||F||60||Non insulin dependent||Left||Severe NPDR||15||36|
|3||M||59||Non insulin dependent||Right||Moderate NPDR||12||18|
|4||F||65||Non insulin dependent||Right||PDR||15||21|
|5||F||55||Non insulin dependent||Right||Moderate NPDR||7||16|
|6||M||38||Non insulin dependent||Left||Moderate NPDR||5||8|
|7||F||51||Non insulin dependent||Right||Severe NPDR||10||12|
|8||M||55||Non insulin dependent||Right||Severe NPDR||10||26|
|9||F||65||Non insulin dependent||Left||Severe NPDR||5||36|
|10||F||55||Non insulin dependent||Left||Severe NPDR||7||24|
|11||M||65||Non insulin dependent||Left||Moderate NPDR||12||16|
|12||F||45||Non insulin dependent||Right||Severe NPDR||6||24|
|13||M||40||Non insulin dependent||Left||Moderate NPDR||12||16|
|14||F||55||Non insulin dependent||Left||Moderate NPDR||15||12|
|15||F||51||Non insulin dependent||Right||Severe NPDR||10||36|
|16||F||60||Non insulin dependent||Right||PDR||13||24|
|17||M||45||Non insulin dependent||Right||Moderate NPDR||5||26|
|18||M||55||Non insulin dependent||Left||Moderate NPDR||3||24|
|19||M||38||Non insulin dependent||Left||Moderate NPDR||3||16|
|20||F||50||Non insulin dependent||Left||Severe NPDR||5||8|
|21||F||58||Non insulin dependent||Left||Severe NPDR||10||24|
|22||M||45||Non insulin dependent||Right||Moderate NPDR||7||26|
|23||M||50||Non insulin dependent||Right||Moderate NPDR||12||24|
|24||F||38||Non insulin dependent||Left||Severe NPDR||10||12|
NPDR = nonproliferative diabetic retinopathy
PDR = proliferative diabetic retinopathy
Table 2: Visual evoked potential and intraocular pressure evaluation (pre- and post-treatment)
|Pat. #||Visual Evoked Potential||Intraocular pressure|
|Latency (msec)||Amplitude (µV)||(mmHg)|
|Initial||1 m||3 ms||6 ms||Initial||1 m||3 ms||6 ms||Initial||1 m||3 ms||6 ms|
Table 3: Evaluation of visual acuity and Optical Cohedence Tomography (pre- and post-treatment)
|Pat. #||Visual acuity||Optical Cohedence Tomography (OCT)|
|Central thickness (m)||Total volume (mm3)|
|Initial||1 m||3 ms||6 ms||Initial||1 m||3 ms||6 ms||Initial||1 ms||3 ms||6 ms|
Table 4: Results of intravitreal triamcinolone injection (mean ± SD; P-value)
|Assessment||Initial||1 month||3 months||6 months|
|VEP (P100), msec|
|Mean ± SD||99.93±2.5||99.51±4.04||100.13±9.14||95.53±4.47|
|Mean ± SD||98.0±2.1|
|P-value||P1: NS||P2: NS||P2: NS||P2:NS|
|VEP (amplitude), µV|
|Mean ± SD||7.31±2.23||15.37±3.88||26.69±3.72||25.65±2.28|
|Mean ± SD||18.1±2.5|
|P-value||P1 < 0.001||P2 < 0.001||P2 < 0.001||P2<0.001|
|Central thickness, µm|
|Mean ± SD||375±35.50||233.33±40.17||145.83±27.58||202.46±29.60|
|Mean ± SD||170.35±15.20|
|P-value||P1< 0.001||P2 < 0.001||P2 < 0.001||P2 < 0.001|
|Macular Volume, mm3|
|Mean ± SD||10.39±1.87||7.73±1.01||6.55±0.99||6.61±1.09|
|Mean ± SD||5.85±1.50|
|P-value||P1< 0.001||P2 < 0.001||P2 < 0.01||P2<0.01|
|Intraocular pressure, mmHg|
|Mean ± SD||13.92±3.85||20.58±8.42||15.83±4.15||14.63±1.48|
|P-value||P1 < 0.001||P2 < 0.001||P2 < 0.001||P2: NS|
P1: initial vs. control values.
P2: one, three, six months follow up results vs. initial values
NS: non significant