Article Text

Research paper
Different neurological and physiological profiles in POEMS syndrome and chronic inflammatory demyelinating polyneuropathy
  1. Saiko Nasu,
  2. Sonoko Misawa,
  3. Yukari Sekiguchi,
  4. Kazumoto Shibuya,
  5. Kazuaki Kanai,
  6. Yumi Fujimaki,
  7. Shigeki Ohmori,
  8. Satsuki Mitsuma,
  9. Shunsuke Koga,
  10. Satoshi Kuwabara
  1. Department of Neurology, Graduate School of Medicine, Chiba University, Chiba, Japan
  1. Correspondence to Dr S Kuwabara, Department of Neurology, Graduate School of Medicine, Chiba University, 1-8-1 Inohana, Chuo-ku, Chiba 260-8670, Japan; kuwabara-s{at}


Background POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein and skin changes) syndrome, a rare cause of demyelinating neuropathy associated with multiorgan involvement, has been increasingly recognised. Polyneuropathy is often an initial manifestation and therefore the disorder can be misdiagnosed as chronic inflammatory demyelinating polyneuropathy (CIDP).

Objective To elucidate whether POEMS syndrome and CIDP are differentiated based on profiles of neuropathy.

Methods Clinical and electrophysiological data were reviewed in consecutive POEMS syndrome (n=51) and typical CIDP (n=46) patients in a single Japanese hospital between 2000 and 2010.

Results Both POEMS and CIDP patients showed symmetric polyneuropathy, physiological evidence of demyelination (70% of POEMS patients fulfilled the electrodiagnostic criteria for definite CIDP) and albuminocytological dissociation; 49% of the POEMS syndrome patients had neuropathy onset and 60% of them were initially diagnosed as having CIDP by neurologists. Clinically, POEMS neuropathy more frequently showed severe leg pain (76% vs 7%; p<0.001), muscle atrophy (52% vs 24%; p=0.005) and distal dominant muscle weakness. Electrophysiologically, POEMS syndrome was characterised by less prolonged distal motor latency (mean 5.6 ms vs 8.1 ms; p<0.001) and higher terminal latency index (0.42 vs 0.33; p=0.006) in the median nerves, and unrecordable tibial and sural responses (p<0.001), suggesting demyelination predominant in the nerve trunk rather than in the distal nerve terminals, and axonal loss in the lower limb nerves.

Conclusions Before development of typical systemic manifestations, POEMS neuropathy can be distinguished from CIDP by the clinical profile and patterns of nerve conduction abnormalities. Recognition of these features leads to early diagnosis and proper treatment for POEMS syndrome.

Statistics from


POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein and skin changes) syndrome is a rare cause of demyelinating neuropathy associated with plasma cell dyscrasia and multiorgan involvement.1–3 Serum levels of vascular endothelial growth factor (VEGF) are elevated in POEMS syndrome, and increased vascular permeability and neovasculisation mediated by VEGF are considered to cause characteristic symptoms.4 POEMS syndrome has been increasingly recognised, and since 2000, over 400 reports have been published from Western and Asian countries.

POEMS syndrome is a serious disease with neurological disability due to progressive polyneuropathy,5 with high mortality by multiorgan failure.6 7 There are no randomised controlled trials for POEMS syndrome8 but several case series have shown benefits of treatments such as high dose chemotherapy with autologous peripheral blood stem cell transplantation9 10 and thalidomide therapy.11

In the advanced stage of POEMS syndrome, the diagnosis is not difficult; the specific combination of polyneuropathy, peripheral oedema, pleural effusion/ascites and skin changes would lead to the diagnosis that can be confirmed by the presence of M protein and elevated serum VEGF levels. However, approximately 50% of POEMS patients initially showed only polyneuropathy.5 Because POEMS syndrome causes peripheral nerve demyelination, the electrodiagnostic findings may be similar to those of chronic inflammatory demyelinating polyneuropathy (CIDP). Previous studies suggest that patterns of nerve conduction abnormalities are different in the two disorders12 13 but these studies included a small number of POEMS patients (eight and 12, respectively). Therefore, the findings are not conclusive and need to be confirmed.

Assuming that patients with POEMS syndrome and those with CIDP require fundamentally different treatments, the differential diagnosis is clinically important. In the past 10 years, we have experienced more than 10 POEMS patients who were initially misdiagnosed as suffering from CIDP by neurologists and received immunoglobulin therapy, resulting in no effects and progression of neuropathy. This led us to investigate whether POEMS syndrome is associated with particular patterns of clinical features and nerve conduction study results, and whether such profiles differ from those of CIDP.



This study included 51 consecutive patients with POEMS syndrome (35 men and 16 women) seen at Chiba University Hospital (Chiba, Japan) between 2000 and 2010. Age ranged from 34 to 73 years (median 53 years). All patients fulfilled the diagnostic criteria of POEMS syndrome,8 and had increased serum VEGF levels measured by ELISA. In the same study period, there were 46 patients with typical CIDP (29 men and 17 women; median age 44 years) whose condition fulfilled the diagnostic criteria for typical CIDP proposed by the European Federation of Neurological Societies and Peripheral Nerve Society (EFNS/PNS).14 Clinical and electrophysiological data of both patient groups were analysed. A total 73 healthy subjects served as normal controls in nerve conduction studies. The study protocol was approved by the ethics committee of Chiba University School of Medicine.


Nerve conduction studies were performed using conventional procedures and by a Viking four electromyography machine. Motor nerve studies were made of the median, ulnar and tibial nerves, including F wave analyses. Terminal latency index (TLI) was calculated using the following formula:

TLI = terminal distance (mm)/(distal latency (ms) × conduction velocity (m/s)). Partial motor conduction block was defined as >50% reduction of compound muscle action potentials between the stimulus sites.14 Antidromic sensory nerve conduction studies were performed in the median and sural nerves.

Statistical analyses

Differences in median values were tested using the Mann–Whitney U test. We compared positive and negative groups using Fisher's exact test for categorical outcomes. Receiver operating characteristic analyses were performed to calculate sensitivity and specificity. A p value of <0.05 was considered to be statistically significant. All statistical analyses were performed using the SAS software program, V.9.2 (SAS Institute Inc).


Clinical features

Of the 51 patients with POEMS syndrome, 25 (49%) showed polyneuropathy as an initial and isolated symptom, and in the remaining patients, foot oedema, skin pigmentation or gynaecomastia preceded neuropathy. In the former group with neuropathy onset, 15 (60%) were initially diagnosed as having CIDP by neurologists and received immunoglobulin treatment. Because of the lack of beneficial effects of immunoglobulin, they were referred to our hospital. The median delay to the correct diagnosis was 12 months, and during this period neurological disability worsened (median Hughes grade by 1; range 0–2), and other multiorgan involvement such as massive pleural effusion and cardiac failure developed in 10 patients.

Table 1 shows the clinical profiles of patients with POEMS syndrome and those with typical CIDP. The two patient groups showed symmetric polyneuropathy. POEMS patients showed distal dominant muscle weakness with amyotrophy, which was mainly seen in the lower extremities, whereas in CIDP patients proximal as well as distal muscles were involved and both upper and lower extremities were almost equally involved. Positive sensory symptoms were significantly different; POEMS patients had neuropathic pain in the distal lower limbs much more frequently, and this substantially disturbed the patients' quality of life. A few CIDP patients complained of pain that was generally mild.

Table 1

Clinical profiles of POEMS syndrome and CIDP patients


Using the EFNS/PNS electrodiagnostic criteria, 70% of the POEMS syndrome patients fulfilled the criteria for definite CIDP. Table 2 shows the results of nerve conduction studies. There were major differences, and compared with CIDP patients, POEMS patients showed less prolonged distal latency (p<0.0001) and higher TLI (p=0.006) in the median nerves, and more frequently unrecordable tibial compound muscle action potentials (56% vs 7%; p<0.0001). Receiver operating characteristic analyses showed a sensitivity of 0.85 and specificity of 0.67 for distal motor latency, and sensitivity of 0.93 and specificity of 0.62 for TLI. Apparent increases in distal compound muscle action potential duration (>9 ms) was found for 6% of POEMS patients and for 13% of CIDP patients. Motor nerve conduction velocities were similarly reduced in both patient groups. In sensory nerve studies, patients with POEMS syndrome had relatively preserved amplitudes of median sensory nerve action potentials but more severe involvement of sural sensory potentials (absent response 68% vs 36%; p=0.002).

Table 2

Nerve conduction data for POEMS syndrome and CIDP patients

Overall, patients with POEMS syndrome had nerve conduction slowing predominant in the nerve trunk, rather than in the distal nerve terminals, and axonal loss was prominent in the lower limb nerves. In contrast, CIDP patients showed conduction slowing in the distal nerve segments and multifocal conduction blocks. Representative waveforms of compound muscle action potentials in the median nerve of a patient with POEMS syndrome and in a patient with CIDP are shown in figure 1.

Figure 1

Findings in median motor nerve conduction studies in a patient with POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein and skin changes) syndrome and in a patient with chronic inflammatory demyelinating polyneuropathy (CIDP). Compound muscle action potentials (CMAPs) were recorded from the abductor pollicis brevis after stimulation at the wrist, elbow and axilla. A 58-year-old man with POEMS syndrome showed normal distal latency (3.6 ms; normal <4.1 ms) but prominently slowed nerve conduction velocity (32 m/s; normal >48 m/s); demyelinative conduction slowing was dominant in the nerve trunk compared with the distal nerve terminals. In the CIDP patient (a 40-year-old woman), distal latency was prominently prolonged (6.3 ms) and distally evoked CMAP amplitude was markedly reduced (0.4 mV; normal >5.0 mV), suggestive of distal conduction block. Conduction block in the forearm segment (60% amplitude reduction) was also observed.


Our results show that approximately 50% of POEMS syndrome patients have neuropathy onset and 60% were misdiagnosed as suffering CIDP, resulting in a significant delay in diagnosis and consequently progression of the disease. Secondly, compared with CIDP, neuropathy in POEMS syndrome is characterised by distal dominant painful polyneuropathy, particularly prominent neuropathic foot pain, being much more frequent than in CIDP. Finally, our findings confirm previous results12 13; different pattern of nerve conduction abnormalities between POEMS syndrome and CIDP, presumably reflecting the distinct pathophysiology of the two disorders.

Currently, POEMS syndrome is an under recognised disease, presumably because of the rarity of the disorder. A national survey conducted in Japan in 2003 showed a prevalence of approximately 0.3 per 100 000,15 and the disorder was considered to be less prevalent in Western countries. The prevalence of CIDP has been reported to be similar in Europe and Japan, ranging from 1.9 to 3.6 per 100 000.16–19 As mentioned above, POEMS syndrome has been increasingly recognised even in Western countries, and a series from the USA published in 2003 included 99 patients.3 It is possible that POEMS syndrome is underestimated. As shown in the present study, a considerable number of patients with POEMS syndrome could be misdiagnosed or even undiagnosed. We suggest that POEMS syndrome should be considered as a differential diagnosis of demyelinating polyneuropathies.

This study also shows that neuropathic pain is a feature of POEMS syndrome. A histological study has reported that pain seen in patients with POEMS syndrome is closely related to a reduction in myelinated, but not unmyelinated, fibre population, suggesting that the painful symptoms in POEMS syndrome may be generated by well preserved unmyelinated C fibres due to lack of inhibitory myelinated A fibres.20

Previous electrophysiological studies have shown that patients with POEMS syndrome have: (1) slowing of nerve conduction that was more predominant in the intermediate than in distal nerve segments; (2) rare conduction block; and (3) more severe attenuation of compound muscle action potentials in the lower than upper limbs.12 13 In contrast, findings in CIDP patients were characterised by conduction slowing predominant in the distal nerve terminals, frequent conduction block in the nerve trunk and less discrepancy between upper and lower limb nerves.12 Our results confirm these observations in a large number of patients. We suggest that, particularly in neuropathy onset POEMS patents without obvious systemic symptoms/signs, the specific patterns of nerve conduction abnormalities indicate the possibility of POEMS syndrome.

In immune mediated neuropathies such as typical CIDP, demyelination preferentially affects specific regions where the blood–nerve barrier is anatomically deficient, namely the distal nerve terminals and nerve roots.21–23 In contrast, in POEMS neuropathy demyelination appears to predominantly involve the nerve trunk where the blood–nerve barrier functions. This is possibly mediated by VEGF induced diffuse breakdown of the blood–nerve barrier and this could result in homogenous rather than focally accentuated demyelination in CIDP.4 23 The distribution of demyelinating lesions implies the pathophysiology of each disorder. POEMS syndrome could be distinguished from CIDP by clinical profile and patterns of nerve conduction abnormalities based on neuropathy features alone. If there are findings suggestive of the POEMS syndrome pattern, systemic and laboratory investigations should lead to the diagnosis. Recognition of these features contributes to early diagnosis and efficacious treatments in POEMS syndrome.


View Abstract


  • See Editorial commentary, p 474

  • Funding This work was supported in part by the Health and Labour Sciences Research Grant on Intractable Diseases (neuroimmunological diseases) from the Ministry of Health, Labour and Welfare of Japan (SK).

  • Competing interests None.

  • Ethics approval Ethics approval was provided by the ethics committee of Chiba University School of Medicine, Chiba, Japan.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles