Purpose
To determine the statistical correlation between visual acuity (VA) and various quantitative parameters relevant to birdshot retinochoroidopathy (BRC) evaluation.
Design
Hospital-based retrospective observational study.
Methods
setting : Institutional. study population : Consecutive HLA29+ BRC patients were included between May and August 2013 at a single tertiary center (Pitié-Salpétrière Hospital, Paris). observation procedures : Demographic data and quantitative parameters relevant to BRC at baseline were collected: VA, degree of anterior and posterior inflammatory reaction, foveal thickness measured by optical coherence tomography (OCT), Arden ratio, and electrooculography (EOG) light peak. main outcome measures : Correlation between VA and the other parameters of the ipsilateral and fellow eye was performed using Spearman rank correlation coefficients.
Results
Fifty-five patients were included. Mean VA was 6/9.5 in the right eye (OD) and 6/12 in the left eye (OS). Mean foveal thickness was 240 μm OD (range: 112–606) and 251 μm OS (range: 85–662). Mean Arden ratio was 159% OD and 160% OS. EOG light peak was 714 mV OD (range: 316–1379) and 746 mV OS (range: 272–1652). VA of a given eye was moderately correlated with VA of the contralateral eye (r = 0.4). On the contrary, all other parameters showed a strong correlation between both eyes (all r > 0.7, P < .01). Overall, none of the studied parameters was correlated with its VA (all r < 0.5).
Conclusion
In BRC, visual acuity alone does not seem to fully reflect the disease severity in terms of clinical or ancillary quantitative findings at baseline.
Birdshot retinochoroidopathy (BRC) is a chronic bilateral posterior form of autoimmune uveitis, characterized by the presence of distinctive hypopigmented choroidal lesions. Although BRC is a potentially blinding pathology, evolution toward retinal atrophy and blindness occurs in a subclinical fashion. As a consequence, disease evaluation and monitoring should not be neglected and must be as complete and objective as possible, including both clinical and ancillary tests.
Among the ancillary parameters, automated visual fields (AVF) and electroretinograms (ERG), especially the 30 Hz ERGs, have shown their reliability and efficiency as evaluation and monitoring tools. However, ophthalmologists continue to experience difficulty in managing patients with BRC owing to the variability and labor-intensive methodology of AVF and ERG in detecting subtle change in disease activity, as opposed to large changes (ie, significant loss of functional vision).
While many BRC specialists agree on the fact that treatment algorithms based on visual acuity (VA), vitreous inflammatory reaction, or retinal vascular leakage of fluorescein alone are ineffective and that many uveitis experts do not consider VA to be an optimal descriptive mechanism for activity in BRC, such statements have never been statistically investigated. This is particularly relevant for VA that seems to be widely used by the nonspecialists as a decisional tool.
This study was designed to determine the statistical correlation between VA and various clinical and ancillary quantitative parameters relevant to BRC evaluation.
Methods
Institutional review board approvals for retrospective chart reviews were obtained commensurate with the respective institutional requirements prior to the beginning of the study. Described research was approved by the Ethics Committee of the French Society of Ophthalmology and adhered to the tenets of the Declaration of Helsinki. Fully informed consent was obtained for all patients.
This was a hospital-based retrospective study that reviewed the files of all consecutive HLA-A29-positive BRC patients seen for the latest time for a routine visit between May and August 2013 at a single tertiary referral center (Pitié Salpétrière Hospital, Paris). All patients met criteria for diagnosis of birdshot retinochoroidopathy that had been previously defined by an international group of uveitis specialists. Additionally, exhaustive evaluation to exclude the infectious and other inflammatory causes of uveitis that can mimic BRC was performed for all patients: Parameters including history and clinical presentation, specific infectious serologies, pulmonary function tests, salivary glandular biopsy, targeted ancillary tests, and clinical examination by an internal medicine specialist were systematically performed to exclude differential diagnoses.
For each patient, demographic data including age, sex, ethnicity, and medical history were recorded at baseline. Quantitative parameters relevant to BRC were collected at baseline for both eyes and included the following clinical parameters: best-corrected visual acuity (BCVA) measured with a decimal scale, then converted to a Snellen scale; quantification of anterior segment cells and vitreous inflammatory reaction; and the following ancillary parameters: central foveal macular thickness as measured by optical coherence tomography (OCT), quantitative electrophysiological data including Arden ratio and electrooculography (EOG) light peak; and automated perimetry data (mean deviation). Anterior segment cells were quantified using a laser cell flaremeter analyzer (Kowa FC 1000; Kowa, Tokyo, Japan) (normal value <8 photon units/ms). Vitreous inflammatory reaction was clinically determined based on the Standardization of Uveitis Nomenclature criteria. OCT analyses were performed using a Cirrus device (Carl Zeiss Meditec, Inc, Jena, Germany). Electrooculograms integrating the protocols recommended by the International Society for Clinical Electrophysiology of Vision and were obtained using a WIN 8000F monitor (Metrovision, Perenchies, France). Automated perimetry was performed with the Humphrey visual field analyzer, using the 24-2 and 10-2 programs (Zeiss-Humphrey, San Leandro, California, USA).
All outcomes were from a single clinical visit (the first visit in our unit). Correlation between VA and the other quantitative parameters of the same eye was analyzed.
Statistical Analyses
Correlations between each pair of quantitative parameters were estimated using Spearman rank correlation coefficients. Each correlation coefficient was tested against the null hypothesis of absence of correlation (ρ = 0). P values of .05 or less were considered statistically significant. Correlations were considered to be strong when correlation coefficients were >0.5 and very strong when correlation coefficients were >0.8. Statistical analyses were performed using R 3.0.2 (R Development core team, 2013, Vienna, Austria).
Results
Fifty-five HLA-A29-positive BRC patients were included. Mean disease duration at baseline was 2.8 years (range: 0.1–24.4 years). Female-to-male sex ratio was 1.6. Demographic and clinical data at baseline are listed in Table 1 . At baseline, 36 of 55 patients (66.6%) were naive of any treatment. 16 of 55 patients had previously received oral corticosteroids while only 3 of 55 (5.4%) had received immunosuppressive treatments (oral cyclosporine, intravenous immunoglobulins, and oral mycophenolate mofetil, respectively). Mean BCVA was 6/9.5 in the right eye (OD) (range: 6/126–6/6) and 6/12 in the left eye (OS) (range: 6/600–6/6). Mean anterior chamber flare values reached 18 photon units (ph)/ms OD (range: 0.1–157 ph/ms) and 23 ph/ms OS (range: 0.1–373 ph/ms). Most patients had moderate vitreous inflammatory reaction (mean 1.2 +, range: 0–3). Mean OCT central foveal thickness was 240 μm OD (range: 112–606 μm) and 251 μm OS (range: 85–662 μm). Mean Arden ratio was 159% OD (range: 106%–218%) and 160% OS (range: 103%–235%), whereas the normal value is >180%. Mean EOG light peak was 714 mV OD (range: 316–1379 mV) and 746 mV OS (range: 272–1652 mV), showing frequent EOG impairment.
| Demographic Characteristics and Patient History | Mean | 95% CI | SD |
|---|---|---|---|
| Patient age (y) | 38.7 | [13–83] | 15.8 |
| Female/male sex ratio | 1.6 | – | – |
| Disease duration at baseline (y) | 2.8 | [0.1–24.4] | 4.7 |
| Medical therapy before and at baseline | |||
| Oral corticosteroids | 29% | – | – |
| Immunosuppressive treatments | 5.5% | – | – |
| Clinical characteristics | |||
| Visual acuity a OD | 6/9.5 | [6/126–6/6] | 6/10 |
| Visual acuity a OS | 6/12 | [6/600–6/6] | 6/15 |
| Anterior chamber flare OD (ph/ms) | 18 | [0.1–157] | 31 |
| Anterior chamber flare OS (ph/ms) | 23 | [0.1–373] | 68 |
| Vitreous inflammatory reaction b OD | 1.2 | [0–3] | 0.6 |
| Vitreous inflammatory reaction b OS | 1.2 | [0–3] | 0.6 |
| OCT central foveal thickness OD (μm) | 240 | [112–606] | 115 |
| OCT central foveal thickness OS (μm) | 251 | [85–662] | 125 |
| Arden ratio OD | 159% | [106–218] | 32 |
| Arden ratio OS | 160% | [103–235] | 37 |
| EOG light peak OD (mV) | 714 | [316–1379] | 319 |
| EOG light peak OS (mV) | 746 | [272–1652] | 346 |
| 24-2 c MD OD | −7.723 | [−29.18 to 11.9] | 7.2 |
| 24-2 c MD OS | −7.923 | [−27.96 to 10.3] | 7.2 |
| 10-2 c MD OD | −4.726 | [−19.7 to 7.3] | 6.3 |
| 10-2 c MD OS | −5.951 | [−14.7 to −0.14] | 4 |
a Visual acuity was measured on a decimal scale and then converted to a Snellen scale.
b Vitreous inflammatory reaction was clinically determined based on the Standardization of Uveitis Nomenclature criteria.
c Automated perimetry performed with the Humphrey visual field analyzer, using the 24-2 and 10-2 programs.
Table 2 shows the 2-by-2 correlation coefficients between each pair of parameters. VA did not seem to be significantly correlated with any other parameter of the study (all correlation coefficients <0.5). Additionally, except for visual acuity, all quantitative parameters related to a given eye seemed strongly correlated with the same parameter in the fellow eye (eg, flare on the right eye with flare on the left eye: r = 0.821, P < .001).
| VA a OD | VA a OS | AH Flare OD | AH Flare OS | Vitreous Inflam b OD | Vitreous Inflam b OS | OCT CFT OD | OCT CFT OS | Arden OD | Arden OS | EOG LP OD | EOG LP OS | 10-2 MD c OD | 10-2 MD c OS | 24-2 MD c OD | 24-2 MD c OS | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VA a OD | —- | 0.48 | 0.41 d | 0.24 | −0.03 d | −0.06 | 0.27 d | 0.14 | −0.17 d | −0.35 | −0.20 d | −0.03 | −0.07 d | 0.06 | −0.07 d | −0.03 |
| VA a OS | <.001 | —- | 0.08 | −0.08 d | −0.28 | −0.27 d | 0.03 | 0.10 d | −0.18 | −0.40 d | 0.17 | 0.27 d | −0.20 | −0.08 d | −0.34 | −0.39 d |
| AH Flare OD | .03 | .68 | —- | 0.82 | 0.00 | 0.10 | 0.24 | 0.38 | −0.22 | −0.13 | 0.29 | 0.00 | −0.31 | 0.16 | −0.10 | 0.15 |
| AH Flare OS | .20 | .69 | <.001 | —- | 0.02 | 0.18 | 0.22 | 0.28 | −0.16 | −0.04 | 0.00 | −0.17 | −0.03 | −0.01 | 0.14 | 0.19 |
| Vitreous Inflam b OD | .85 | .04 | .98 | .91 | —- | 0.89 | −0.06 | −0.04 | −0.34 | −0.22 | 0.16 | 0.09 | −0.14 | −0.20 | −0.07 | −0.03 |
| Vitreous Inflam b OS | .65 | .05 | .61 | .36 | <.001 | —- | −0.15 | −0.09 | −0.30 | −0.19 | 0.14 | 0.02 | −0.25 | −0.40 | −0.23 | −0.14 |
| OCT CFT OD | .07 | .87 | .25 | .29 | .70 | .32 | —- | 0.77 | 0.04 | 0.06 | −0.16 | 0.08 | −0.08 | 0.17 | 0.34 | 0.29 |
| OCT CFT OS | .36 | .52 | .07 | .18 | .77 | .57 | <.001 | —- | −0.01 | 0.12 | 0.04 | 0.19 | 0.08 | 0.42 | 0.42 | 0.42 |
| Arden OD | .41 | .38 | .48 | .62 | .10 | .15 | .85 | .95 | —- | 0.72 | 0.25 | 0.11 | 0.60 | 0.20 | 0.36 | 0.39 |
| Arden OS | .08 | .05 | .70 | .91 | .29 | .36 | .79 | .59 | <.001 | —- | 0.19 | 0.16 | 0.30 | 0.00 | 0.57 | 0.73 |
| EOG LP OD | .39 | .49 | .50 | >.999 | .49 | .56 | .53 | .89 | .28 | .43 | —- | 0.91 | −0.40 | −0.40 | 0.67 | 0.62 |
| EOG LP OS | .90 | .25 | >.999 | .69 | .71 | .92 | .76 | .46 | .65 | .50 | <.001 | —- | −0.80 | −0.80 | 0.82 | 0.69 |
| 10-2 MD c OD | .77 | .39 | .27 | .92 | .54 | .29 | .75 | .73 | .35 | .68 | .75 | .33 | —- | 0.57 | 0.50 | 0.25 |
| 10-2 MD c OS | .80 | .74 | .56 | .97 | .37 | .07 | .47 | .06 | .78 | >.999 | .75 | .33 | .01 | —- | 0.49 | 0.63 |
| 24-2 MD c OD | .67 | .03 | .65 | .52 | .64 | .15 | .03 | .01 | .16 | .02 | .02 | .00 | .02 | .02 | —- | 0.82 |
| 24-2 MD c OS | .87 | .01 | .49 | .35 | .87 | .38 | .08 | .01 | .12 | .00 | .03 | .01 | .27 | .00 | <.001 | —- |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
