R J Blanch1,2,3, J Horsburgh4, A Creavin5, M A Burdon4, C Williams5. 1. Academic Department of Military Surgery and Trauma, Royal Centre for Defence Medicine, Birmingham, UK. blanchrj@bham.ac.uk. 2. Neuroscience and Ophthalmology, Institute of Inflammation and Ageing, University of Birmingham, Birmingham, UK. blanchrj@bham.ac.uk. 3. Birmingham Neuro-Ophthalmology, University Hospital Birmingham NHS Trust, Birmingham, B15 2WB, UK. blanchrj@bham.ac.uk. 4. Birmingham Neuro-Ophthalmology, University Hospital Birmingham NHS Trust, Birmingham, B15 2WB, UK. 5. Bristol Medical School: Population Health Sciences, University of Bristol, Bristol, UK.
Abstract
BACKGROUND: Overdiagnosis of papilloedema is common and carries significant potential for morbidity from over-investigation and over-treatment. We aimed to determine the community prevalence of false positive diagnosis of papilloedema (FPE) on fundus imaging. METHODS: We evaluated fundus images from a community cross-section of 198 12-14-year-olds from the Avon Longitudinal Study of Parents and Children (ALSPAC) longitudinal cohort study database and patient images from our hospital departmental database with and without papilloedema. We asked clinicians, in isolation, to rate the subjects as a forced choice task to "papilloedema" or "not papilloedema" based on the fundus images alone. Raters comprised (i) four neuro-ophthalmologists, (ii) four ophthalmologists, (iii) four neurologists and (iv) four emergency medicine physicians. RESULTS: The prevalence of FPE in the ALSPAC population, defined as images mistaken as papilloedema by χ% of raters (Pχ) varied from P100 = 0% to P50 = 21.3 ± 3.9%. In the hospital population, there was a lower rate of FPE, P50 = 7.1 ± 10.8%. Sensitivity for papilloedema detection approached 100%, though three raters incorrectly labelled the same patient with unilateral disc swelling as normal, all other cases were detected by all raters. CONCLUSIONS: Fundus photography assessment in isolation is highly sensitive but poorly specific for papilloedema detection. Using this method to screen the general population has significant potential for harm as overdiagnosis occurs, even in the hands of experienced clinicians.
BACKGROUND: Overdiagnosis of papilloedema is common and carries significant potential for morbidity from over-investigation and over-treatment. We aimed to determine the community prevalence of false positive diagnosis of papilloedema (FPE) on fundus imaging. METHODS: We evaluated fundus images from a community cross-section of 198 12-14-year-olds from the Avon Longitudinal Study of Parents and Children (ALSPAC) longitudinal cohort study database and patient images from our hospital departmental database with and without papilloedema. We asked clinicians, in isolation, to rate the subjects as a forced choice task to "papilloedema" or "not papilloedema" based on the fundus images alone. Raters comprised (i) four neuro-ophthalmologists, (ii) four ophthalmologists, (iii) four neurologists and (iv) four emergency medicine physicians. RESULTS: The prevalence of FPE in the ALSPAC population, defined as images mistaken as papilloedema by χ% of raters (Pχ) varied from P100 = 0% to P50 = 21.3 ± 3.9%. In the hospital population, there was a lower rate of FPE, P50 = 7.1 ± 10.8%. Sensitivity for papilloedema detection approached 100%, though three raters incorrectly labelled the same patient with unilateral disc swelling as normal, all other cases were detected by all raters. CONCLUSIONS: Fundus photography assessment in isolation is highly sensitive but poorly specific for papilloedema detection. Using this method to screen the general population has significant potential for harm as overdiagnosis occurs, even in the hands of experienced clinicians.