PATIENTS AND METHODS: Clinical characteristics and electrophysiological data of sixty-one consecutive patients admitted between 2012 and 2015 were systematically analysed and reclassified according to the new GBS clinical classification. Neurophysiology was evaluated with Hadden et al.'s vs recently proposed Rajabally et al.'s criteria. Functional severity and clinical outcome of various GBS subtypes were ascertained.
RESULTS: All patients initially identified as GBS or related disorders can be sub-classified into having classical GBS (41, 67%), classic Miller-Fisher Syndrome (MFS) (6, 10%), Pharyngeal-cervical-brachial (PCB) (3, 5%), paraparetic GBS (4, 7%), bifacial weakness with paresthesia (3, 5%), acute ophthalmoparesis (AO) (1, 2%) and overlap syndrome (3, 5%): one (2%) with GBS/Bickerstaff brainstem encephalitis overlap and 2 (3%) with GBS/MFS overlap. Greater proportion of axonal classical GBS (67% vs 55%, p=0.372) seen with Rajabally et al.'s criteria and a predominantly axonal form of paraparetic variant (75%) independent of electrodiagnostic criteria were more representative of Asian GBS cohort. Classical GBS patients had lowest admission and discharge Medical Research Council Sum Score (MRCSS), greater functional disability and longest length of in-patient stay. Twenty (20/21, 95%) patients who needed mechanical ventilation had classical GBS. Patients required repeated dose of intravenous immunoglobulin (5/6, 3%) or plasma exchange (4/4, 100%) more frequently had axonal form of classical GBS.
CONCLUSION: Phenotype recognition based on new GBS clinical classification, supported by electrodiagnostic study permits more precise clinical subtypes determination and outcome prognostication.
METHODS: Albuminocytologic dissociation (ACD) was defined as an increased protein level (>0.45 g/L) in the absence of elevated white cell count (<50 cells/μL). We excluded 124 (8%) patients because of other diagnoses, protocol violation, or insufficient data. The CSF was examined in 1,231 patients (89%).
RESULTS: In 846 (70%) patients, CSF examination showed ACD, which increased with time from weakness onset: ≤4 days 57%, >4 days 84%. High CSF protein levels were associated with a demyelinating subtype, proximal or global muscle weakness, and a reduced likelihood of being able to run at week 2 (odds ratio [OR] 0.42, 95% CI 0.25-0.70; p = 0.001) and week 4 (OR 0.44, 95% CI 0.27-0.72; p = 0.001). Patients with the Miller Fisher syndrome, distal predominant weakness, and normal or equivocal nerve conduction studies were more likely to have lower CSF protein levels. CSF cell count was <5 cells/μL in 1,005 patients (83%), 5-49 cells/μL in 200 patients (16%), and ≥50 cells/μL in 13 patients (1%).
DISCUSSION: ACD is a common finding in GBS, but normal protein levels do not exclude this diagnosis. High CSF protein level is associated with an early severe disease course and a demyelinating subtype. Elevated CSF cell count, rarely ≥50 cells/μL, is compatible with GBS after a thorough exclusion of alternative diagnoses.
CLASSIFICATION OF EVIDENCE: This study provides Class IV evidence that CSF ACD (defined by the Brighton Collaboration) is common in patients with GBS.