Ari Breiner1,2, Pierre R Bourque1,2, Jeffrey A Allen3,4. 1. Department of Medicine, Division of Neurology, The Ottawa Hospital, University of Ottawa, 1053 Carling Ave, Room ES-08, Ottawa, M5G 2C4, Ontario, Canada. 2. Ottawa Hospital Research Institute, Ottawa, Ontario, Canada. 3. Department of Neurology, Northwestern University, Chicago, Illinois, USA. 4. Department of Neurology, University of Minnesota, Minneapolis, Minnesota, USA.
Abstract
INTRODUCTION: Recent literature has concluded that cerebrospinal fluid total protein (CSF-TP) upper reference limits (URL) should be higher than 45 mg/dl and stratified by age. METHODS: Data-driven URLs were applied to the analysis of a cohort of patients with correctly and incorrectly diagnosed chronic inflammatory demyelinating polyneuropathy (CIDP). Descriptive statistics were calculated, and exploratory analyses were used to test the impact of different CSF-TP URLs on sensitivity and specificity of CIDP diagnosis. RESULTS: The adoption of higher and age-dependent CSF-TP URLs reduced the sensitivity of CSF analysis slightly (from 95% to 84%-86%); however, the overall CIDP detection rate was unchanged. Twelve of 36 (33%) false-positive diagnoses occurred with CSF-TP elevation as the sole supportive criteria. By applying updated CSF-TP URLs, the specificity of CSF analysis increased from 39% to 57%-64%. DISCUSSION: Implementation of data-driven CSF-TP URLs improves CIDP diagnostic specificity without compromising sensitivity, thereby lessening CIDP misdiagnosis. Muscle Nerve 60: 180-183, 2019.
INTRODUCTION: Recent literature has concluded that cerebrospinal fluid total protein (CSF-TP) upper reference limits (URL) should be higher than 45 mg/dl and stratified by age. METHODS: Data-driven URLs were applied to the analysis of a cohort of patients with correctly and incorrectly diagnosed chronic inflammatory demyelinating polyneuropathy (CIDP). Descriptive statistics were calculated, and exploratory analyses were used to test the impact of different CSF-TP URLs on sensitivity and specificity of CIDP diagnosis. RESULTS: The adoption of higher and age-dependent CSF-TP URLs reduced the sensitivity of CSF analysis slightly (from 95% to 84%-86%); however, the overall CIDP detection rate was unchanged. Twelve of 36 (33%) false-positive diagnoses occurred with CSF-TP elevation as the sole supportive criteria. By applying updated CSF-TP URLs, the specificity of CSF analysis increased from 39% to 57%-64%. DISCUSSION: Implementation of data-driven CSF-TP URLs improves CIDP diagnostic specificity without compromising sensitivity, thereby lessening CIDP misdiagnosis. Muscle Nerve 60: 180-183, 2019.