Portal de Periódicos da CAPES

Comparison of retinal thickness and fundus-related microperimetry with visual acuity in uveitic macular oedema

2009; Wiley; Volume: 89; Issue: 6 Linguagem: Inglês

10.1111/j.1755-3768.2009.01750.x

1755-3768

Martin Roesel, B. Heimes, Carsten Heinz, Andreas Henschel, Georg Spital, Arnd Heiligenhaus,

Retinal and Optic Conditions

Purpose: Macular oedema is a common complication and vision-limiting factor in uveitis. The aim of this study was to compare retinal thickness as measured by optical coherence tomography and photoreceptor function as measured by fundus-related microperimetry with respect to their correlation with visual acuity. Methods: Prospective observational monocentre study. Thirty-one patients (53 eyes) with endogenous uveitis and fluorescein angiographically confirmed macular oedema were evaluated. Foveal thickness was analysed using spectral-domain (SpectralisTM; Heidelberg Engineering, Heidelberg, Germany) OCT and retinal sensitivity was assessed using fundus-related microperimetry (MP1; Nidek Technologies, Padova, Italy). All findings were correlated with best-corrected visual acuity (BCVA). Results: Foveal thickness was correlated with BCVA [p = 0.005, r = 0.38, 95% confidence interval (CI) 0.12–0.59]. For microperimetry measurements, a negative correlation with logMAR visual acuity was found. Fixation abnormalities were not associated with poor visual acuity, increased foveal thickness or retinal sensitivity. In eyes with cystoid changes in the outer plexiform and inner nuclear layer, foveal thickness was increased (p < 0.0001). Epiretinal membrane formation was present in 70%. In these eyes, foveal thickness was significantly increased (p = 0.003) and visual acuity was worse (p = 0.08). Conclusion: Foveal thickness and fundus-related microperimetry were correlated with visual acuity. Cystoid changes in the outer plexiform and inner nuclear layer and the presence of epiretinal membrane were associated with poor visual acuity. Fixation abnormalities were not associated with poor visual acuity. Macular oedema represents a major complication of uveitis and impairs visual acuity in many patients (Rothova et al. 1996; Durrani et al. 2004). Optical coherence tomography (OCT) is a non-invasive, objective method of measuring retinal thickness that appears to be suitable for detecting and monitoring uveitic macular oedema (Reinthal et al. 2004). This technique has also contributed to our understanding of the different types of uveitic macular oedema (Markomichelakis et al. 2004) and may give information on their visual prognosis (Sivaprasad et al. 2007). The evidence for a correlation between foveal thickness and visual acuity is conflicting: although several authors have observed a good correlation in patients with diabetic macular swelling (Hee et al. 1995, 1998; Sanchez-Tocino et al. 2002), this issue remains controversial for uveitic macular oedema (Antcliff et al. 2000; Sivaprasad et al. 2007). Recently, a new OCT generation with spectral-domain technology (SD-OCT) was developed, offering better image resolution, three-dimensional imaging and faster imaging speed. In a recent study, the advantages of the new high-resolution spectral-domain OCT could be observed in evaluating the macular pathological features of a series of patients with uveitis (Gupta et al. 2008). Microperimetry is a useful tool for assessing macular sensitivity and is even able to quantify photoreceptor function. Retinal function is assessed by using microperimetry in relation to the fundus, and thus spatial light increment sensitivity can be mapped. Moreover, the patient’s fixation is examined. Both microperimetry and OCT show advantages over multifocal electroretinography, which is another psychophysical tool for the assessment of macular function. Microperimetry has already been evaluated for diabetic macular oedema, idiopathic macular teleangiectasia and age-related macular degeneration (Rohrschneider et al. 1997, 2000, 2001; Okada et al. 2006; Carpineto et al. 2007; Charbel et al. 2007, 2008). Using a laser scanning ophthalmoscope without an eye-tracking device, Lardenoye et al. (2000) reported decreased sensitivity in the macular area, but only five eyes with inflammatory oedema were included in their study. Vujosevic et al. (2006) concluded that microperimetry may be of value in predicting the outcome of diabetic macular oedema. The aim of the present study was to correlate macular sensitivity and foveal thickness with visual acuity, and to assess fixation abnormalities in patients with uveitic macular oedema. Fifty-three eyes of 31 patients were included in this prospective, cross-sectional, single-centre study. All patients suffered from uveitis with clinical evidence of macular oedema. The presence of cystoid macular oedema /(CMO) was confirmed by fluorescein angiography (FA). Perifoveal leakage of the dye that accumulated in cystic spaces in a petaloid pattern and showed diffuse positive staining at the fovea in the late frames was diagnosed as macular oedema. Twenty eyes had 31 had diffuse oedema, and two others had additional subretinal neurosensory detachment. All patients were examined for any associated disease, and best-corrected visual acuity (BCVA) was assessed. Slit-lamp examination, Goldmann tonometry and ophthalmoscopy were performed to detect any uveitis-related complications. The diagnostic work-up included complete differential blood count with differential count, Lyme titres and western blot, fluorescent treponemal antibody absorption, chest X-ray, angiotensin-converting enzyme, antinuclear antibodies (ANA), tuberculosis skin test, human leucocyte antigen (HLA)-B27, creatinine, liver enzymes, blood coagulation and magnetic resonance imaging of the brain. The patients were examined by a rheumatologist, neurologist, dermatologist and ear, nose and throat consultant. Patients with significant cataract or vitreous opacities were not included in this study. OCT imaging was performed with a spectral-domain combined scanning laser angiography/OCT device (Spectralis OCT; Heidelberg Engineering, Heidelberg, Germany). The Spectralis system utilized the Heidelberg Eye Explorer Software 1.5.12 with the Spectralis viewing module 3.1.0. The scans were acquired by an experienced examiner and with the pupils dilated. Correct foveal placement was controlled by the infrared fundus image captured in parallel. A single horizontal line scan passing through the fovea was taken for analysis. Retinal thickness was measured automatically using the built-in software routines. The minimum foveal thickness value was used for further statistical analysis. Results were evaluated independently by two experienced observers. In cases of elapsed foveal depression, localization of the fovea was supported by the infrared image. The fundus-related microperimeter [MP1; Nidek Technologies, Padova, Italy (software version 1.6.0)] was used to measure macular sensitivity. After dilating the pupils (1% tropicamide and 5% phenylephrine), a reference frame was obtained with the integrated infrared camera. We used a standardized macular scan pattern with 4-2-1 double staircase test strategy. Then, 49 test stimuli with a Goldmann III stimulus size, and a projection time of 200 ms were projected onto the central 16°. A standard red cross with 5° diameter was used for fixation and an automated eye-tracking system was used to compensate fundus movements. White background illumination was 4 apostilbs, and stimulus intensity was 0–20 dB. The mean sensitivity values at the five central locations, representing the central 5° of the visual field, the nine locations in the horizontal line (horizontal 16°) and the central point (central 2°) were included in the analysis (Fig. 1). The fundus movements were tracked during examination while the patient watched the fixation target. Microperimetry patterns for retinal sensitivity measurement. The measurements of the central 2° (ellipse), horizontal 16° (rectangle) and central 5° (polygon) were used for analysis. In accordance with Fujii et al. (2003), the fixation pattern was graded on fixation location and fixation stability. Fixation stability was documented in three categories: stable, relatively unstable or unstable. Fixation location was classified as predominantly central, poorly central or predominantly eccentric. BCVA was measured and transformed into logarithm of the minimum angle of resolution (logMAR) for statistical analysis. This study followed the tenets of the Declaration of Helsinki. Informed consent was obtained from all patients before they were included in the study. Data were analysed using MedCalc® version 9.3 (MedCalc Software, Mariakerke, Belgium). Normal data distribution was checked by using the Kolmogorov–Smirnov test. Fisher’s exact test for categorical data and t-test were applied for statistical analysis when appropriate. Pearson’s correlation coefficient (r) was used for data correlation. p < 0.05 was judged to be statistically significant. Fifty-three eyes of 31 patients (nine male and 22 female) were included. Mean age was 51 years [standard deviation (SD) 14 years]. An associated disease was found in 13 patients. Sarcoidosis, multiple sclerosis and Behcet’s disease were the most common diseases. Eleven patients had anterior uveitis, ten intermediate uveitis, two posterior uveitis and eight panuveitis. Mean BCVA (logMAR) was 0.41 (SD 0.32). Epidemiological data are shown in Table 1. The correlations of visual acuity to foveal thickness and microperimetry (Pearson’s correlation coefficient) are summarized in Table 2. Foveal thickness correlated with BCVA (p = 0.005, r = 0.38, 95% CI 0.12–0.59, Pearson’s correlation coefficient). CMO duration did not correlate with visual acuity (p = 0.183). In eyes with cystoid changes in the outer plexiform and inner nuclear layer, foveal thickness was increased (p < 0.0001, unpaired t-test). Retinal sensitivity correlated negatively with foveal thickness for the central 2° (p = 0.006, r = −0.41), the horizontal 16° (p = 0.001, r = −0.48) and the central 5° (p = 0.002, r = −0.47; Pearson’s correlation coefficient). A significant negative correlation was found between microperimetry measurement and visual acuity for the central 2°, the central 5° and the horizontal 16° (Table 2). Fixation was stable in 47%, relatively unstable in 44% and unstable in 9%. Neither visual acuity (p = 0.834, unpaired t-test) nor foveal thickness (p = 0.069, unpaired t-test) differed between patients with stable fixation and patients with relatively unstable or unstable fixation. For retinal sensitivity (dB), no difference was observed between eyes with normal or abnormal fixation (central point p = 0.752, central five points p = 0.760, horizontal nine points p = 0.969; unpaired t-test). Fixation stability did not differ in eyes with cystoid changes in the outer plexiform or inner nuclear layer compared with eyes without cystoid changes (p = 0.46, Fisher’s exact test). Location of fixation was predominantly central in 79%, relatively eccentric in 14% and predominantly eccentric in 7%. For location of fixation, there were no differences (unpaired t-test) in visual acuity (p = 0.471), foveal thickness (p = 0.992) and retinal sensitivity (central point p = 0.280, central five points p = 0.714, horizontal nine points p = 0.947). Epiretinal membrane formation was present in 70%. In these eyes, foveal thickness was significantly increased (p = 0.003) and visual acuity (p = 0.08) was worse than in normal eyes. Mean CMO duration was 21 months. Spearman’s coefficient of rank correlation (rho) was r = 0.381 (95% CI 0.124–0.591) for CMO duration and foveal thickness. For the correlation between CMO duration and retinal sensitivity, rho was −0.291 (95% CI −0.544 to 0.00972) for the central 2°, rho = −0.286 (95% CI −0.540 to 0.0157) for the horizontal 16° and rho = −0.121 (95% CI −0.540 to 0.0157) for the central 5°. Because macular oedema is a severe vision-threatening complication of uveitis (Lardenoye et al. 2006), detecting and monitoring this condition appears to be of particular importance. The underlying pathogenic mechanism for the development of macular oedema is not yet understood clearly. Inflammatory cell infiltration and protein secretion, retinal cell damage, vitreoretinal traction and epiretinal membrane formation may all represent contributing factors. The blood–retinal barrier may break down as a result of retinal pigment epithelium (RPE) changes, capillary endothelium damage or loss of autoregulatory control of blood flow (Freeman et al. 2001). Dick (1994) reported that after immune-mediated damage occurred, the vessels behave as if chronically damaged. This article highlights the association between OCT and fundus-related microperimetry to visual acuity in patients with uveitis and macular oedema. The gold standard for detecting macular oedema is FA, but the sensitivity of OCT for identifying it was found to be comparable (Antcliff et al. 2000). One of the main advantages of OCT over FA is that side-effects are virtually absent because of the non-invasive nature of the procedure; OCT also takes less time and provides an objective thickness measurement. Furthermore, OCT provides additional information about the morphology of the vitreoretinal interface and can depict tractional membranes. OCT can also help to distinguish between different patterns of uveitic macular oedema (Markomichelakis et al. 2004). In a recent study, Sivaprasad et al. (2007) evaluated the response of uveitic macular oedema to medical treatment by OCT and found that inner CMO is more resistant to treatment than other types of oedema. In the present study, foveal thickness was measured using a spectral-domain OCT that quantifies this from the inner limiting membrane to the outer margin of the outer highly reflective layer, which may correspond to the RPE-choriocapillaris complex. A critical point for comparability of different studies is the use of different landmarks for analysing foveal thickness by employing different OCT devices. The correlation between foveal thickness and visual acuity was described as moderate to strong for diabetic macular swelling (Hee et al. 1998; Otani et al. 1999). No correlation was found for macular teleangiectasia (Charbel et al. 2008). For uveitic macular oedema, controversy still exists about whether a relation exists between foveal thickness and visual acuity. Whereas Reinthal et al. (2004) did not find a relation between foveal thickness and visual acuity, Antcliff et al. (2000) observed a weak correlation. Sivaprasad et al. (2007) reported an inverse correlation (r = 0.55) between macular thickness and logMAR visual acuity. In our study, we also found a negative correlation between logMAR BCVA and foveal thickness. Using spectral-domain OCT, epiretinal membrane formation was observed in 70% of eyes in our patient group. Foveal thickness in these eyes was greater than in eyes without an epiretinal membrane. Cystoid changes in the outer plexiform and inner nuclear layer were associated with increased foveal thickness and worse visual acuity. For retinal sensitivity as detected by fundus-related microperimetry, we found a significant negative correlation to logMAR visual acuity overall. In contrast to the OCT measurement, cystoid retinal changes or epiretinal membrane growth were not associated with poor retinal sensitivity. Our findings are in line with the results of Lardenoye et al. (2000), but that study included only five patients with inflammatory macular oedema and the examination was performed using a scanning laser ophthalmoscope (SLO) device without eyetracking, which reduces the quality and consistency of the measurement. The correlation of retinal sensitivity to foveal thickness as detected with fundus-related microperimetry has not yet been proven in uveitic patients. In our study, we found a negative correlation. Fundus-related microperimetry offers additional information about the function of the central visual field and the stability and pattern of fixation. In patients with diabetic macular oedema, Carpineto et al. (2007) reported poor visual acuity and retinal sensitivity and found increased foveal thickness in patients with fixation impairment. For uveitic oedema, fixation had not yet been evaluated. In contrast to the results of Carpineto et al., we found no association between fixation stability or fixation location and visual acuity, retinal sensitivity or foveal thickness in this series of uveitis patients. Whereas fixation was eccentric in 47.65% of Carpineto et al.’s diabetic population (relatively or predominantly eccentric), we found eccentric fixation in only 21% of our uveitis population. These differences (p = 0.004, Fisher’s exact test) are probably the result of the different methods used to diagnose macular oedema. Whereas the oedema was detected clinically in Carpineto’s study, it was detected by FA in ours. Our study could have contained a higher number of eyes with subclinical macular oedema. Furthermore, hard exudates centred at the foveola and intraretinal bleedings could have caused fixation alterations in diabetic patients. Another possible explanation is the difference in the underlying pathological mechanisms, with a potentially different pattern of retinal/choriocapillaris damage. In terms of stability of fixation, our results do not differ significantly from those in the cited diabetic group: Carpineto et al. found unstable (relatively unstable or unstable) fixation in 59.5% whereas in our group of uveitic eyes fixation was unstable in 53% (p = 0.57, Fisher’s exact test) (Carpineto et al. 2007). Again, for unstable fixation associated with visual acuity, the increased foveal thickness or retinal sensitivity as measured by fundus-related microperimetry failed to reach the level of significance. Both foveal thickness and fundus-related microperimetry correlate with visual acuity. Whereas OCT discloses further information about the vitreoretinal anatomy, fixation ability can be assessed by fundus-related microperimetry. Cystoid retinal changes and epiretinal membrane growth were associated with poor visual acuity. Clinical trial registration: this study is registered at http://www.clinicaltrials.gov, registration number NCT00791726.