Logistic regression models to predict CIDP with six clinical features and sural nerve biopsy in a group of 64 patients in whom CIDP was considered in the differential diagnosis
| Clinical feature | Model 1: after entering six clinical features | Model 2: after entering the significant (p⩽0.20) clinical features, as identified in the first model, and sural nerve biopsy | ||||
|---|---|---|---|---|---|---|
| OR | P Value3-150 | OR | P Value3-150 | |||
| (1) Remitting course | 1.6 | 0.70 | — | — | ||
| (2) Symmetric sensorimotor neuropathy | 5.3 | 0.14 | 5.0 | 0.15 | ||
| (3) Areflexia | 1.3 | 0.80 | — | — | ||
| (4) Raised CSF protein concentration (0.5–1.0 g/l) | 7.7 | 0.13 | 9.0 | 0.11 | ||
| Highly raised (>1 g/l) | 38.5 | 0.01 | 58.1 | 0.01 | ||
| (5) Neurophysiological studies consistent with CIDP | 51.7 | 0.003 | 61.8 | 0.002 | ||
| (6) Absence of comorbidity | 31.5 | 0.03 | 36.6 | 0.02 | ||
| (7) Sural nerve biopsy consistent with CIDP | — | — | 2.2 | 0.42 | ||
Dashes indicate that these factors were not entered into the model
↵3-150 Calculated with Wald statistics.