Premium
O1–09–02: Beyond interest and limits of CSF biomarkers in clinical practice: The PLM study
Author(s) -
Gabelle Audrey,
Dumurgier Julien,
Vercruysse Olivier,
Paquet Claire,
Bombois Stephanie,
Laplanche Jean Louis,
Peoch Katell,
Schraen Susanna,
Buee Luc,
Pasquier Florence,
Hugon Jacques,
Touchon Jacques,
Lehmann Sylvain
Publication year - 2013
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2013.04.099
Subject(s) - receiver operating characteristic , medicine , logistic regression , biomarker , memory clinic , alzheimer's disease , medical physics , medical diagnosis , multidisciplinary approach , disease , oncology , dementia , pathology , social science , biochemistry , chemistry , sociology
Background: We aimed to assess the impact of CSF biomarkers on diagnosis and management of memory clinic patients in the daily practice. In addition we took into account the considerations and views of the neurologists when using CSF biomarkers. Methods: All patients visiting the VUmc Alzheimer center for cognitive screening were consecutively included from 06-2011 till 05-2012. Clinical diagnoses were made in a multidisciplinary team without knowledge of the CSF biomarkers. Neurologists filled out questionnaires before and after disclosure of CSF results, with change of diagnosis and confidence in diagnosis as primary outcome measures. In addition, neurologists expressed their diagnostic considerations, reasons for wishing to know the CSF biomarker results, interpretation of the biomarker concentrations and impact on patient management. Results: 438 patients were included, of which 80% (351 patients) underwent lumbar puncture. Before disclosure of CSF results neurologists wished to know the results in 74% of all cases. Reasons included: confirm (46%) or exclude Alzheimer’s disease (AD) (36%), reassure the patient (11%), assess the prognosis (24%), or preselection for clinical trials (11%). Diagnostic confidence level was higher when neurologists thought not to need CSF results (90% vs 82%, p< 0.001). After disclosure of CSF results, diagnosis changed in 23 cases (7% of patients with CSF). In cases where the diagnosis remained unchanged, confidence level increased from 84 to 89% (p<0.001). The increase was significant in AD (6%, p<0.001), other dementia (4%, p< 0.01) and MCI patients (6%, p< 0.001). In 40 patients (13% of patients with CSF and unchanged diagnosis) level of confidence decreased substantially because CSF was not in accordance with the clinical diagnosis. These cases often underwent more intensive clinical follow-up or further investigations, such as FDGand PiB-PET, or consultation of other specialists. Conclusions: In this study in a tertiary referral setting, CSF biomarkers did not change the diagnosis markedly, but the clinician’s confidence in the underlying neuropathology. In cases where CSF results contradicted clinical diagnosis, they prompted further investigation or follow-up. We conclude that CSF biomarkers may aid clinicians in the diagnostic process, even in a setting where pre-CSF confidence in the diagnosis is high.