Neurophysiological analysis of the cauda equina in POEMS syndrome

Abstract

In POEMS syndrome (Crow-Fukase syndrome), the proximal conduction of peripheral nerves innervating leg muscles is difficult to assess using conventional nerve conduction studies (NCSs) when F-waves or H-reflexes are not detectable. Recently, we developed a novel magnetic stimulation method to measure cauda equina conduction time (CECT) using a powerful coil known as a magnetic augmented translumbosacral stimulation coil (MATS coil) [1–3]. This method allows us to investigate CECT even in patients with peripheral neuropathy [3]. In this study, we applied our method to a patient with POEMS syndrome. The patient was a 72-year-old man who was unable to walk 5 months after onset. Neurological findings at 6 months after onset revealed weakness with muscular atrophy (Medical Research Council scale for muscle strength grades: proximal arm 4, distal arm 3, proximal leg 3 and distal leg 1), distal dominant sensory disturbance and areflexia of the extremities. Whole-body computed tomography revealed hepatomegaly and multiple enlarge lymph nodes. We made the diagnosis of POEMS syndrome on the basis of skin hemangioma, skin pigmentation, hirsutism, leg edema, hepatomegaly, lymphadenopathy, monoclonal protein of immunoglobulin G (k type), and subacute motor and sensory polyneuropathy. Conventional NCSs were performed on the right side. In the median nerve, the amplitude of compound muscle action potentials (CMAPs) was severely reduced to 0.4 mV. Neither a pathological temporal dispersion nor a conduction block was detected. Motor conduction velocity (MCV) between wrist and elbow was slowed to 24 m/s with prolonged distal latency (5.6 ms). Sensory nerve action potentials (SNAPs) were not evoked. For the lower extremities, neither CMAPs in the tibial nerve nor SNAPs in the sural nerve were evoked. Additionally, neither F-waves in the median and tibial nerves nor H-reflexes in the tibial nerve were evoked. Needle electromyography (nEMG) was performed in the biceps brachii, first dorsal interossei, rectus femoris, and tibialis anterior muscles on the right side. Abundant spontaneous activities were recorded during relaxation in only the first dorsal interossei and tibialis anterior muscles. In all the contracted muscles, recruitment of motor unit potentials was reduced during voluntary contraction (contraction of the tibialis anterior muscle was impossible). MATS coil stimulation was performed to measure CECT. CMAPs were recorded from the biceps femoris muscle. This particular proximal muscle was selected since it had experienced much less atrophy than the distal muscles. The most distal part of the cauda equina was activated by MATS coil stimulation over the first sacral spinous processes (S1) [1–3]. The most proximal part of the cauda equina was activated by MATS coil stimulation over the first lumbar spinous processes (L1) [2, 3]. As shown in Fig. 1b, this method demonstrated severely prolonged CECT (right 8.1 ms, left 8.0 ms; normal values H. Matsumoto (&) H. Hashida Department of Neurology, Japanese Red Cross Medical Center, 4-1-22 Hiroo, Shibuya-ku, Tokyo150-8935, Japan e-mail: hideyukimatsumoto.jp@gmail.com