Abstract
Purpose:
To evaluate the incidence of fundus changes in patients with pulmonary diseases, in particular pulmonary arterial hypertension using a special ultra-widefield retinal imaging (UWRI,OPTOS 200 Tx, Scotland) in comparison to conventional biomicroscopy (CB).
Methods:
During a period of 6 months we recruited 246 in- and outpatients from the Department of Pneumology of the University Hospital Giessen. The study was approved by the IRB. Images were recorded with undilated pupils and evaluated independently by two consultants (C1 and C2) of the Department of Ophthalmology. CB was done with a slit-lamp (Haag Streit 900, Switzerland) and a 90-D lens in mydriasis by 2 groups of specialists (G1 and G2), independently from one another. Data were analysed for the most frequent pathological fundus changes seen in UWRI images by both consultants and with CB by both groups of specialists. Inter-examiner agreement was assessed separately for each of the pathologies.
Results:
Complete data was acquired by UWRI and by CB in 458 eyes (229 patients), out of whom 82 patients suffered from pulmonary arterial hypertension. The incidence of pathologies such as blurred optic disc margin, optic disc pallor, vessel dilation and vessel tortuosity varied considerably, not only between C1 and C2 for the OPTOS images, but also between G1 and G2 for CB. Overall, macular drusen were detected in 30.6 %, retinal drusen in 24.0%, and retinal degeneration in 29.3% of the 458 eyes. However, when comparing the results for C1, C2, G1 and G2, concordant results were seen in only 3.0% of RE (right eyes) and 1.35% of LE (left eyes) for macular drusen, in 1.85% of RE and 1.79% of LE for retinal drusen, and in 4.2% of RE and 9.52% of LE for retinal degeneration. Concordance was higher between G1 and G2 than between C1 and C2.
Conclusions:
Retinal pathologies were present in a high percentage of patients with pulmonary disease, in particular pulmonary arterial hypertension. However, agreement between C1 and C2 as well as between G1 and G2 was low. Review of the OPTOS images by the same and additional readers will allow validating the results. This is not possible for CB by G1 and G2 who also showed significantly discordant results. Objective documentation therefore still appears desirable but needs further investigations.
Keywords: 459 clinical (human) or epidemiologic studies: biostatistics/epidemiology methodology •
465 clinical (human) or epidemiologic studies: systems/equipment/techniques •
688 retina