Nerve conduction study and nerve ultrasound as biomarkers for steroid dependence in chronic inflammatory demyelinating polyneuropathy

Biomarkers for disease activity are lacking in chronic inflammatory demyelinating polyradiculoneuropathy (CIDP). We aimed to investigate whether motor nerve conduction studies (NCSs) and nerve ultrasound and their follow-up changes could predict steroid-dependency and treatment refractoriness. Sixty-three CIDP patients were followed up with both nerve ultrasound and NCS. Cross-sectional areas (CSAs) were measured on the bilateral median, ulnar nerves and brachial plexus. NCSs were performed on the median and ulnar nerves. Patients with normal or mildly slow MCV at the first visit were less likely to be steroid-dependent and had lower INCAT at the last follow-up (median 0 [0,1]), whereas those with dramatically slow MCV were more likely to be steroid-dependent and had higher INCAT at the last follow-up (median 2[2,2]) (p = 0.009 for steroid dependent, p = 0.004 for INCAT). None of the patients whose MCV improved above the lower normal limit were steroid-dependent, whereas nearly half of those whose MCV decreased or remained unchanged were steroid-dependent (p = 0.005). A two-step method had a sensitivity of 85% and specificity of 80% for distinguishing patients with steroid dependency. First, we divided patients into three groups according to the MCV change. Second, we explored the trend of steroid-dependent and treatment-refractory based on the CSA at admission and change in CSA. For patients whose MCV improved beyond the threshold, the risk of relapse was low, and we suggest more rapid tapering of steroid. For those with decreased MCV, the risk of relapse was greater and slower steroid tapering or immunosuppressant use is suggested.