Article Text

PDF

Research paper
Atypical CIDP: diagnostic criteria, progression and treatment response. Data from the Italian CIDP Database
  1. Pietro Emiliano Doneddu1,
  2. Dario Cocito2,
  3. Fiore Manganelli3,
  4. Raffaella Fazio4,
  5. Chiara Briani5,
  6. Massimiliano Filosto6,
  7. Luana Benedetti7,
  8. Anna Mazzeo8,
  9. Girolama Alessandra Marfia9,
  10. Andrea Cortese10,
  11. Brigida Fierro11,
  12. Stefano Jann12,
  13. Ettore Beghi13,
  14. Angelo Maurizio Clerici14,
  15. Marinella Carpo15,
  16. Angelo Schenone16,
  17. Marco Luigetti17,
  18. Giuseppe Lauria18,19,
  19. Giovanni Antonini20,
  20. Tiziana Rosso21,
  21. Gabriele Siciliano22,
  22. Guido Cavaletti23,
  23. Giuseppe Liberatore1,
  24. Lucio Santoro3,
  25. Erdita Peci2,
  26. Stefano Tronci4,
  27. Marta Ruiz5,
  28. Stefano Cotti Piccinelli6,
  29. Antonio Toscano8,
  30. Giorgia Mataluni9,
  31. Laura Piccolo10,
  32. Giuseppe Cosentino11,
  33. Mario Sabatelli17,24,
  34. Eduardo Nobile-Orazio1,25
  35. on behalf of the Italian CIDP Database study group
    1. 1 Neuromuscular and Neuroimmunology Service, Humanitas Clinical and Research Institute, Rozzano, Italy
    2. 2 Department of Neuroscience, University of Turin, Turin, Italy
    3. 3 Department of Neuroscience, Reproductive Sciences and Odontostomatology, University of Naples ‘Federico II’, Naples, Italy
    4. 4 Division of Neuroscience, Department of Neurology, Institute of Experimental Neurology (INSPE), San Raffaele Scientific Institute, Milan, Italy
    5. 5 Neurology Unit, Department of Neuroscience, University of Padua, Padua, Italy
    6. 6 Center for Neuromuscular Diseases and Neuropathies, Unit of Neurology, ASST ‘Spedali Civili’, University of Brescia, Brescia, Italy
    7. 7 Neurology Unit, Sant’Andrea Hospital, La Spezia, Italy
    8. 8 Department of Clinical and Experimental Medicine, Unit of Neurology, University of Messina, Messina, Italy
    9. 9 Dysimmune Neuropathies Unit, Department of Systems Medicine, Tor Vergata University of Rome, Rome, Italy
    10. 10 IRCCS Foundation C. Mondino National Neurological Institute, Pavia, Italy
    11. 11 Department of Experimental BioMedicine and Clinical Neurosciences (BioNeC), University of Palermo, Palermo, Italy
    12. 12 Department of Neuroscience, Niguarda Ca’ Granda Hospital, Milan, Italy
    13. 13 Laboratorio di Malattie Neurologiche, IRCCS-Istituto Mario Negri, Milan, Italy
    14. 14 Neurology Unit, Circolo and Macchi Foundation Hospital, Insubria University, DBSV, Varese, Italy
    15. 15 ASST Bergamo Ovest-Ospedale Treviglio, Treviglio, Italy
    16. 16 Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa and IRCCS AOU San Martino-IST, Genoa, Italy
    17. 17 Unit of Neurology, IRCCS Foundation Policlinico A. Gemelli, Rome, Italy
    18. 18 Unit of Neuroalgology, IRCCS Foundation ‘Carlo Besta’ Neurological Institute, Milan, Italy
    19. 19 Department of Biomedical and Clinical Sciences ‘Luigi Sacco’, University of Milan, Milan, Italy
    20. 20 Unit of Neuromuscular Diseases, Department of Neurology Mental Health and Sensory Organs (NESMOS), Faculty of Medicine and Psychology, ‘Sapienza’ University of Rome, Sant’Andrea Hospital, Rome, Italy
    21. 21 ULSS2 Marca Trevigiana, UOC Neurologia-Castelfranco Veneto, Treviso, Italy
    22. 22 Neurology Unit, Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy
    23. 23 School of Medicine and Surgery and Experimental Neurology Unit, University of Milano-Bicocca, Monza, Italy
    24. 24 NEuroMuscular Omnicentre (NEMO), Serena Onlus, Foundation-Pol. A. Gemelli, Rome, Italy
    25. 25 Department of Medical Biotechnology and Translational Medicine, University of Milan, Milan, Italy
    1. Correspondence to Professor Eduardo Nobile-Orazio, Neuromuscular and Neuroimmunology Service, IRCCS Humanitas Research Hospital, Rozzano 20089, Italy; eduardo.nobile{at}unimi.it

    Abstract

    Objectives A few variants of chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) have been described, but their frequency and evolution to typical CIDP remain unclear. To determine the frequency and characteristics of the CIDP variants, their possible evolution to typical CIDP, and treatment response.

    Methods We applied a set of diagnostic criteria to 460 patients included in a database of Italian patients with CIDP. Clinical characteristics and treatment response were reviewed for each patient. The Kaplan-Meier curve was used to estimate the progression rate from atypical to typical CIDP.

    Results At the time of inclusion, 376 (82%) patients had a diagnosis of typical CIDP while 84 (18%) had atypical CIDP, including 34 (7%) with distal acquired demyelinating symmetric neuropathy (DADS), 17 (4%) with purely motor, 17 (4%) with Lewis-Sumner syndrome (LSS) and 16 (3.5%) with purely sensory CIDP. Based on retrospective review of the symptoms and signs present at onset and for at least 1 year, 180 (39%) patients had an initial diagnosis compatible with atypical CIDP that in 96 (53%) patients evolved to typical CIDP. Mean disease duration was longer in patients evolving to typical CIDP than in those not evolving (p=0.0016). Patients with DADS and LSS had a less frequent response to immunoglobulin than those with typical CIDP, while patients with purely motor and sensory CIDP had a similar treatment response.

    Conclusions The proportion of patients with atypical CIDP varies during the disease course. DADS and LSS have a less frequent response to intravenous immunoglobulin compared with typical CIDP, raising the possibility of a different underlying pathogenetic mechanism.

    • chronic inflammatory demyelinating polyradiculoneuropathy
    • lewis–sumner syndrome
    • distal acquired demyelinating symmetric neuropathy
    • CIDP
    • diagnostic criteria

    Statistics from Altmetric.com

    Introduction

    Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a chronic and disabling immune-mediated polyradiculoneuropathy.1 Several clinical variants of CIDP have been reported widening the spectrum of this neuropathy. These variants have been defined by the Joint Task Force of the European Federation of Neurological Societies and the Peripheral Nerve Society (EFNS/PNS) as ‘atypical CIDP’, and include distal acquired demyelinating symmetric neuropathy (DADS), purely motor or sensory CIDP, Lewis-Sumner syndrome (LSS) and focal CIDP.1 The diagnostic criteria for these forms are however not well defined possibly explaining their variable frequency ranging from 1% to 49% in different series2 3 and the reported differences in their treatment response.4–7 It is also unclear the frequency and time to their possible evolution to typical CIDP.5 8

    We established a set of diagnostic criteria for the CIDP variants derived from a revision of the literature and applied it to a large series of patients with CIDP included in a web-based database on Italian patients with CIDP, to determine the frequency and characteristics of the CIDP variants, their possible evolution to typical CIDP, and their treatment response.

    Patients and methods

    Database and study population

    From January 2015 to April 2018, we included data from 500 patients with a diagnosis of CIDP or one of its variants followed by 22 Italian Centres with expertise in immune-mediated neuropathies. The diagnosis of CIDP was made by the treating neurologist and reviewed by the coordinating Centre (PED and EN-O) and classified according to the EFNS/PNS diagnostic criteria.1 The reasons for suspecting CIDP when nerve conduction studies were not diagnostic were also reported by the treating neurologist and included, besides a clinical history and presentation consistent with CIDP, abnormality of the supportive tests1 (cerebrospinal fluid (CSF) analysis, ultrasound or MRI of the nerves and plexus, sensory conduction studies or somatosensory evoked potentials, nerve biopsy and response to previous therapy) and a relapsing course of the disease. All the data were included by the treating neurologist in a web-based electronic database expressly prepared by CINECA, Bologna, Italy.

    Diagnosis of atypical CIDP

    We reviewed the literature on the diagnostic criteria used for atypical CIDP4–29 (online supplementary table 1) and defined a set of clinical diagnostic criteria for all the patients included in our study (box 1). Our criteria include mandatory and exclusion criteria for each atypical CIDP form while symptoms and signs not essential for the diagnosis (but that may be part of the clinical picture) were defined as ‘other possible symptoms and signs’. We also decided that a minimum of 1-year duration of symptoms and signs specific to each atypical form was necessary to establish a diagnosis of atypical CIDP. This is because even typical CIDP may initially present with purely sensory or motor symptoms evolving over a few months to a typical sensorimotor form. We defined purely sensory CIDP as a non–length-dependent sensory neuropathy, DADS as a length-dependent sensory or sensorimotor neuropathy. Since, however, some authors included length-dependent purely sensory neuropathy as sensory CIDP, we also separately analysed the data of patients with sensory or sensorimotor DADS. We defined LSS as a sensory or sensorimotor multifocal neuropathy or in the presence of a clearly distinct degree of impairment among contiguous nerves. We included under this form focal CIDP where symptoms were homogeneously restricted to the nerves of one limb or two limbs (ipsilateral upper and lower limb). For chronic immune sensory polyradiculopathy (CISP), we used the criteria proposed by Sinnreich and coworkers.9 Depending on the distribution of symptoms, patients with only sensory impairment without weakness were therefore split in our study into pure sensory CIDP including CISP, pure sensory DADS or pure sensory LSS. We adhered to the criteria of the EFNS/PNS1 for the diagnosis of typical CIDP, but we also included under this diagnosis patients with bilateral, although asymmetric, but not multifocal, motor and sensory impairment (at least one Medical Research Council (MRC) point difference between the two sides).

    Box 1

    Proposed clinical diagnostic criteria for atypical CIDP

    Mandatory criteria

    (A) with or (B) without increased distal motor latency

    • Symmetric, sensory or sensorimotor symptoms and signs starting distally in the lower limbs, without proximal limb–trunk–face involvement (length-dependent fashion).*

    Other possible symptoms

    • Ataxia, neuropathic pain, cramps, fatigue, autonomic symptoms, tremor.

    • Upper-limb distal sensory or sensorimotor symptoms and signs occurring later (at least after 1 year from onset).

    Exclusion criteria

    • Cranial nerve involvement.

    • Proximal limbs, trunk, face involvement.

    • Weakness without sensory symptoms.

    • Symptoms and signs starting in the upper limbs.

    Mandatory criteria

    (A) with or (B) without abnormal motor nerve conduction studies

    • Sensory symptoms (including ataxia), without weakness, in a polyneuropathic distribution, symmetric or asymmetric.*

    • Symptoms may start anywhere in the body excluding a length-dependent pattern (included under DADS).

    Other possible symptoms

    • Neuropathic pain, fatigue, tremor.

    • Facial sensory symptoms.

    Exclusion criteria

    • Motor symptoms/signs including cramps and motor cranial nerve palsy.

    • Multifocal distribution.

    • Autonomic dysfunction.

    Mandatory criteria

    • Sensory symptoms with a polyneuropathic distribution without weakness.*

    • Normal motor and sensory nerve conduction and EMG studies.

    PLUS at least two of the following:

    • Abnormal SSEP not due to CNS involvement.

    • MRI showing gadolinium enhancement and/or hypertrophy of the cauda equina, lumbosacral or cervical nerve roots, or the brachial or lumbosacral plexuses.

    • Elevated CSF protein level with normal cells.

    Exclusion criteria and other possible symptoms

    • As in pure sensory CIDP.

    Mandatory criteria

    (A) with or (B) without abnormal sensory nerve conduction studies

    • Weakness, without sensory symptoms or signs, in a polyneuropathic distribution, symmetric or asymmetric.*

    • Symptoms may start anywhere in the body.

    Other possible symptoms

    • Cramps, fatigue, tremor.

    • Motor cranial nerve palsy.

    Exclusion criteria

    • Sensory symptoms/signs including sensory ataxia.

    • Autonomic dysfunction.

    • Neuropathic pain.

    • Multifocal distribution.

    Mandatory criteria

    (A) with or (B) without motor conduction block

    • Sensory symptoms, with or without weakness, in a multifocal distribution (unilateral focal† CIDP included).*

    • Symptoms may start anywhere in the body.

    Other possible symptoms

    • Cramps, fatigue, autonomic symptoms, ataxia, neuropathic pain.

    • Motor and/or sensory cranial nerve palsy.

    Exclusion criteria

    • Weakness in isolation, without sensory symptoms.

    • Symptoms/signs in a polyneuropathic distribution.

    • *Clinical phenotype must have lasted at least 1 year (temporal criterion).

    • †Focal CIDP defined as the presence of symptoms homogeneously restricted to the nerves of one limb or two limbs (ipsilateral upper and lower limb).

    • CIDP, chronic inflammatory demyelinating polyradiculoneuropathy; CISP, chronic immune sensory polyradiculopathy; CNS, central nervous system; CSF, cerebrospinal fluid; DADS, distal acquired demyelinating symmetric neuropathy; EMG, electromyography; LSS, Lewis-Sumner syndrome; SSEP, somatosensory evoked potential.

    Some authors also included additional specific electrodiagnostic parameters for the diagnosis of atypical CIDP (online supplementary table 1) even if these were not considered in the EFNS/PNS Guidelines for CIDP. These parameters include abnormally increased motor distal latency (DL) in patients with DADS, normal sensory nerve conduction studies in patients with purely motor CIDP, normal motor nerve conduction studies in sensory CIDP and presence of conduction blocks (CBs) in LSS. In all our patients with typical and atypical CIDP, we used the electrodiagnostic criteria of the EFNS/PNS1 for the diagnosis of CIDP, but we subclassified the patients into atypical CIDP based on clinical criteria to allow a retrospective analysis of the diagnosis at onset and of its evolution during the course of the disease. We compared, however, the data of patients within each group who fulfilled or not the additional electrodiagnostic parameters for the diagnosis of atypical CIDP at the time of inclusion in the study. Since the data on distal distance were not available, we did not calculate the terminal latency index in patients with DADS.

    Clinical assessment and ancillary tests

    At enrolment, all patients underwent a detailed clinical history including time of clinical onset, presence–distribution and date of onset of motor and sensory symptoms, ataxia, pain, tremor, cramps, fatigue, autonomic dysfunction and cranial nerve involvement. This information was integrated with the data reported in the medical records. Disease course was defined as progressive, relapsing or monophasic by the treating neurologist. Response to previously performed therapy were reported by the treating neurologist as ‘improved’, ‘stable’ or ‘worsened’. Since the number of patients with atypical CIDP treated with other therapies besides steroids and intravenous immunoglobulin (IVIg) was relatively small, we focused our analysis on these two therapies. The clinical evaluation at enrolment included assessment of muscle strength using the MRC sum score on 12 muscles (range 0–60). Neurological disability was evaluated at enrolment with the Inflammatory-Rash Overall Built Disability Scale (I-RODS) (range 1–48) and the Inflammatory Neuropathy Cause and Treatment (INCAT) disability scale (range 0–10), while Quality of life (QoL) was assessed with the EuroQol-5D-3L scale (five items, each with a score from 1—best to 3—worst). Results of previously performed examinations, including CSF analysis, nerve ultrasound or brachial/lumbosacral plexus MR examination and sural nerve biopsy, were reported when available. The results of nerve conduction studies performed during the course of the disease were included. Sensory nerve conduction studies were performed bilaterally in median, ulnar and sural nerves and included evaluation of sensory nerve action potential amplitude, DL and sensory conduction velocity. Motor nerve conduction studies were also performed bilaterally in median, ulnar, common peroneal and tibial nerves and included distal and proximal compound muscle action potential amplitude (onset to peak) and duration, motor conduction velocity (MCV), distal and proximal motor latency and in most patients F-wave latency. Abnormalities of motor and sensory nerve conduction studies consistent with demyelination were defined according to the EFNS/PNS criteria.1 Some patients also underwent somatosensory evoked potentials. At the time of inclusion, patients with an alternative diagnosis for the neuropathy or high titres of anti-MAG (myelin-associated glycoprotein) antibodies (over 7000 (BTU) by Bühlman method) or without available nerve conduction studies were excluded. In all the patients, the diagnosis at entry was revised by the coordinating Centre according to the aforementioned diagnostic criteria (box 1). These criteria were also applied at clinical presentation based on symptoms present at onset and for at least 1 year. Progression of atypical to typical CIDP was based on the time of appearance of symptoms before the inclusion in the study. CIDP diagnosis was classified in ‘definite’, ‘probable’ or ‘possible’ according to the EFNS/PNS criteria.1

    Statistical analysis

    Descriptive statistics were reported for the entire sample of patients with CIDP, and for typical and the atypical CIDP forms separately. Categorical variables were described using frequencies and percentages, while continuous variables were described using mean, medians and ranges. Demographic, clinical and neurophysiological features, treatment response, strength deficit, disability level and QoL were compared between different subgroups of patients with the χ2 or Fisher’s exact test for categorical variables, and with the t-test or the Wilcoxon-Mann-Whitney U test for continuous variables. For patients with atypical CIDP at onset, the progression rate from atypical to typical CIDP was calculated using Kaplan-Meier survival curves, considering the different atypical forms separately. All tests were two-tailed and the significance level was set to 0.05. Analyses were performed with SAS V.9.4 (SAS Institute, Cary, North Carolina, USA).

    Results

    Frequency and clinical course of atypical CIDP

    By April 2018, 500 patients were enrolled in the database. Twenty-three patients were excluded from the analysis for the presence of an alternative diagnosis (18 patients with anti-MAG titres ranging from 9300 to over 70 000 BTU (median 41 191 BTU), one with Charcot-Marie Tooth 1A, three with amyloidosis and one with only cranial nerve palsy), and 17 for the absence of available nerve conduction studies or incomplete data. A total 460 patients were included in the study (293 men; 167 women), aged 11–92 years, (mean 58; median 60 years), with mean disease duration of 8.3 years (range 0.5–60 years, median 6 years).

    At study entry, 376 (82%) patients had a diagnosis of typical CIDP and 84 (18%) of atypical CIDP. The diagnosis fulfilled the EFNS/PNS diagnostic criteria for CIDP in 82% of the patients with typical CIDP and in 76% of those with atypical CIDP (p=0.2149). Among the patients with atypical CIDP, 34 (7% of total CIDP patients) had DADS, 17 (4%) had purely motor CIDP, 17 (4 %) had LSS including three with focal CIDP and 16 (3.5%) had purely sensory CIDP including two (0.5%) with CISP (figure 1B). When we retrospectively reviewed the symptoms and signs at disease onset, we found that 180 (39%) patients fulfilled the diagnosis of atypical CIDP. This included 59 (13%) patients with DADS, 49 (11%) with purely sensory CIDP including two with CISP, 40 (9%) with purely motor CIDP, 29 (6%) with LSS and three (1%) with CIDP clinically restricted to cranial nerves (cranial CIDP) (figure 1A). Ninety-six (53%) patients with atypical CIDP at onset progressed to typical CIDP after mean disease duration of 5.5 years (range 1–38; median 3) including 25/59 (42%) with DADS, 33/49 (67%) with purely sensory CIDP, 23/40 (57.5%) with purely motor CIDP, 12/29 (41%) with LSS and three with cranial CIDP. Figure 2 shows the yearly progression rate to typical CIDP for patients with different atypical CIDP forms at onset. The clinical form with the highest progression rate was pure sensory CIDP, while DADS showed the lowest rate. Within 5 years, 48% of patients with sensory CIDP, 32% of pure motor, 36% of LSS and 24% of DADS had progressed to typical CIDP; the corresponding values at 10 years were 77%, 64%, 63% and 39% (figure 2). Patients with atypical CIDP who progressed to typical CIDP had longer disease duration compared with patients who did not progress (mean, 11 vs 7 years; p=0.0016).

    Figure 1

    (A) CIDP type at 1 year after onset of symptoms; (B) CIDP type at study entry (mean 8 years after onset). Graphs show the frequency of the atypical CIDP variants at 1 year after the onset of symptoms (A) and at study entry (B) in 460 patients. CIDP, chronic inflammatory demyelinating polyradiculoneuropathy; DADS, distal acquired demyelinating symmetric neuropathy; LSS, Lewis-Sumner syndrome.

    Figure 2

    Probability of progressing from atypical CIDP to typical CIDP. Graph shows the yearly progression rate to typical CIDP of the individual atypical CIDP forms at onset. CIDP, chronic inflammatory demyelinating polyradiculoneuropathy; DADS, distal acquired demyelinating symmetric neuropathy; LSS, Lewis-Sumner syndrome.

    Clinical and electrophysiological features of the different forms of atypical CIDP

    In table 1 are summarised the clinical features and treatment response of the patients with different forms of atypical CIDP in comparison with patients with typical CIDP.

    Table 1

    Comparison of clinical features and treatment response in atypical and typical CIDP

    Distal acquired demyelinating symmetric neuropathy

    Thirty-four patients had a diagnosis of DADS at study entry including 22 with only sensory symptoms without weakness. All patients had a symmetric presentation of symptoms and signs distally in the lower limbs. Compared with typical CIDP, patients with DADS had an older age at onset (mean 58 vs 49 years; p=0.0032). At onset, 26 (76%) patients presented with purely sensory symptoms while eight (24%) also had ankle weakness at presentation. After a mean disease duration of 8 years (range 0.5–26 years), sensory symptoms had progressed to involve also distal upper limbs in 12 (35%) patients, while 12 (35%) patients had distal weakness that in eight was restricted to the lower limbs. Ataxia was present in 16 (47%) patients, five (15%) had postural and intention tremor in the arms, and 14 (41%) patients had pain. Disease course was chronic-progressive in 18 (53%) patients, relapsing in 14 (41%) patients and monophasic in two patients.

    Twenty-four (70.5%) patients met the EFNS/PNS diagnostic criteria for CIDP (21 definite, three probable). Seven of them had increased motor DL. Even if there were some differences between patients with (seven patients) or without (17 patients) increased motor DL in terms of frequency of upper-limb involvement (28.5% vs 47% patients; p=0.6921) and presence of weakness (14% vs 41% patients; p=0.6417), none of the differences was significant.

    Compared with typical CIDP, patients with DADS had higher MRC sum score (mean 58 vs 53.5; p=0.0003), lower disability levels measured with INCAT (mean 1.5 vs 2.7; p=0.0005) and I-RODS (mean 39 vs 33; p=0.0027), and better QoL (mean 7 vs 8; p=0.0030).

    Purely sensory CIDP

    Sixteen patients had a diagnosis of purely sensory CIDP at study entry including two with CISP. In three patients, sensory symptoms were confined to the upper limbs, in three patients to the proximal and distal areas of the lower limbs, in nine patients to the four limbs, and in one patient to the face–trunk and the four limbs. Strength was normal in all patients. Ataxia and fatigue were reported by three (19%) and four (25%) patients, respectively. The clinical course was relapsing in seven (44%) patients, progressive in eight (50%) patients and monophasic in one.

    Twelve (75%) patients met the EFNS/PNS diagnostic criteria for CIDP (11 definite, one probable) with demyelinating features on motor conduction studies despite the absence of motor symptoms and signs.

    No significant difference between purely sensory and typical CIDP was found in terms of severity assessed with INCAT (mean, 1.7 vs 2.7; p=0.0532) and I-RODS (mean, 38 vs 33; p=0.0799), and QoL (mean, 7 vs 8; p=0.1310).

    Purely motor CIDP

    Seventeen patients had a diagnosis of purely motor CIDP at study entry. All patients presented with weakness without sensory symptoms, initially distributed in the upper limbs in two (12%) patients, in the lower limbs in nine (53%) patients and in the four limbs in six (35%) patients. After mean disease duration of 10 years (0.5–28 years), 12 (70.5%) patients had weakness in the proximal and distal muscles of the four limbs while clinical involvement remained confined to the upper limbs in one (6%) patient and to the lower limbs in four (23.5%) patients. In four (23.5%) patients, clinical manifestations were asymmetric without a multifocal distribution. Sensory examination was normal in all patients. Disease course was relapsing in seven (41%) patients and chronic progressive in 10 (59%) patients.

    Fifteen (88%) patients met the EFNS/PNS criteria of CIDP (14 definite, one possible). The most common electrophysiological findings in these patients were CBs (11 patients) and reduced MCV (nine patients). Sensory nerve conduction studies were normal in only five patients despite the absence of sensory symptoms. There was no significant clinical difference between patients with or without abnormal sensory conduction studies.

    No significant difference between purely motor and typical CIDP was found in terms of severity of motor impairment measured with MRC sum score (mean, 51 vs 53.5; p=0.1792), disability by the INCAT (mean, 3.5 vs 2.7; p=0.0972) and I-RODS (mean, 31 vs 33; p=0.7017), and QoL (mean 8 vs 8; p=0.9260).

    Lewis-Sumner syndrome

    Seventeen patients had a diagnosis of LSS at study entry. Compared with typical CIDP, patients with LSS more frequently were men (p=0.0079). Initial symptoms were exclusively sensory in three (18%) patients and sensorimotor in 14 (82%), with symptoms and signs predominantly confined to one upper limb in four (23.5%), one lower limb in six (35%) and one upper plus one lower limb in four (24%) patients. Three (18)% patients had focal CIDP, with symptoms in only one limb. All but two patients developed sensorimotor disturbances after a mean of 7 years (range 0.5–24 years). The clinical course was relapsing in eight (47%) patients, progressive in six (35%) patients and monophasic in three (18%) patients.

    Thirteen patients (76%) met the EFNS/PNS criteria for CIDP (12 definite, one probable) including 11 with CBs in motor nerves. There was no difference between patients with or without CBs in terms of presence of weakness (90% vs 83%; p=1.0000) and frequency of involvement of upper and lower limbs (44% vs 57%; p=0.6284).

    Patients with LSS had less disability by I-RODS (mean, 40 vs 33; p=0.0301) but not INCAT (mean, 2 vs 2.7; p=0.2063) compared with patients with typical CIDP, with similar MRC values (mean, 57 vs 53.5; p=0.0778) and QoL (mean 7 vs 8; p=0.2453).

    Response to therapy in atypical CIDP

    In table 1 is summarised the response to treatment in patients with atypical CIDP. Response to one or more therapy was reported in 16/25 (64%) treated patients with DADS including 9/15 (60%) fulfilling the EFNS/PNS criteria. There was no difference in treatment response between patients with DADS with sensory or sensorimotor disturbance and between patients with DADS with or without increased motor DL. Response to therapy occurred in 9/10 (90%) treated patients with sensory CIDP including 8/8 (100%) fulfilling the EFNS/PNS criteria. Improvement was also observed in 15/17 (88%) treated patients with purely motor CIDP, including 14/15 (93%) patients fulfilling the EFNS/PNS criteria. There was no difference in the overall response to therapy between patients with purely motor CIDP with or without abnormal sensory nerve conduction studies. Response to steroids was also observed in 3/7 (43%) patients with purely motor CIDP. None of the patients with normal sensory nerve conduction studies improved however after steroid therapy while all improved patients had abnormal sensory conduction studies. Response to therapy was reported in 10/15 (67%) treated patients with LSS, including 7/11 (64%) patients fulfilling the EFNS/PNS criteria. There was no difference between patients with LSS with or without CBs.

    Compared with typical CIDP, the overall response to treatment was significantly lower in patients with DADS (p=0.0050) and LSS (p=0.0433) even when we restricted the analysis to patients fulfilling EFNS/PNS criteria that in all but one patient with purely motor CIDP and two with typical CIDP were consistent with definite or probable CIDP (table 1). The same occurred for patients with purely sensory DADS (p=0.0109) but not for patients with purely sensory (p=1.000) or purely motor CIDP (p=0.7182). Patients with DADS and LSS also had a less frequent response to IVIg (p=0.0178 and p=0.0086) compared with typical CIDP and the same occurred for patients fulfilling EFNS/PNS criteria even if the difference remained significant only for LSS (table 1). A less frequent response to IVIg was also observed in patients with sensory DADS (p=0.0108) but not in patients with pure motor or pure sensory CIDP. There was no difference among the different groups in the response to steroids. Among the patients with typical CIDP, 235 (62.5%) patients had symmetric CIDP while 141 (37.5%) an asymmetric but not multifocal CIDP including 36 (9.5%) with a slight asymmetry (one MRC point difference between the two sides). There was no difference between patients with symmetric or asymmetric typical CIDP in the overall response to therapy (87% vs 87%; p=1.0000) or in the response to IVIg (78% vs 78%; p=1.0000) or steroids (48% vs 56%; p=0.2564) while the difference in the overall response to treatment (p=0.0483) and to IVIg (p=0.0113) was significant between patients with asymmetric typical CIDP and LSS. The difference in the IVIg response remained significant (p=0.0135) between patients with LSS and asymmetric typical CIDP with more than one MRC point difference between the two sides, and close to significance in the overall response to therapy (p=0.0545).

    Discussion

    There is some discrepancy on the frequency of the CIDP variants with numbers ranging from 1% to 49%.2 3 It is unclear whether this discrepancy reflects the use of different diagnostic criteria in previous studies or a different disease duration at the time of ascertainment. There are indeed some differences in the definition of these phenotypes.4–29 We applied our set of diagnostic criteria to all our patients and found that the frequency of atypical CIDP at study entry was 18%, similarly to some previous studies (17.8%–19.6%).10 11 When we retrospectively analysed the symptoms at the time disease onset and for the following year, we found that 39% of the patients had a clinical presentation consistent with atypical CIDP that in 53% of the patients evolved to typical CIDP by the time of inclusion in the study. A similar rate of progression (20%–71%) was previously reported in other small series of patients.5 8 12 In our study, progression to typical CIDP was significantly associated with a longer disease duration. Still, 55% of our patients with atypical CIDP at study entry had disease duration of at least 5 years and 30% of at least 10 years. This finding supports the idea that progression to typical CIDP is often, though not invariably, associated with the disease duration. Some patients with typical CIDP presented some clinical modifications during the course of the disease developing a predominantly distal or sensory but not motor or multifocal impairment. The previous more diffuse or sensorimotor impairment did not lead to a change in their diagnosis. There are also some differences in the reported frequency of the different forms of atypical CIDP. The relatively small prevalence of LSS in our series (4%) compared with some previous series (0.5%–34%)3 6 10 11 13 may reflect the inclusion in our series of only patients with a multifocal neuropathy and not of those with an asymmetric polyneuropathy. This decision was also supported by the absence of difference in the response to therapy between patients with symmetric or asymmetric typical CIDP and by the different response to therapy between patients with LSS and asymmetric CIDP. Similarly, we found a lower prevalence of sensory CIDP (3.5%) compared with previous series (11%–35%).14 15 This may partly reflect our use of stringent diagnostic criteria with the proportion of patients raising to 7.8% if we also included patients with purely sensory DADS or LSS. It is also possible that the disease duration explains the difference with a prevalence of sensory CIDP in our series decreasing from 11% at onset to 3.5% at enrolment. DADS was the most common CIDP variant in our study (7%) (2%–17% in previous studies),2 6 10 followed by the purely motor form (4%) (4%–10% in previous series).2 10 13

    Only few studies have compared disability in typical and atypical CIDP.6 30 31 Two studies found a higher level of disability and impairment in patients with typical CIDP than in patients with LSS and DADS.6 30 Another study did not find a difference in terms of disability between typical and atypical CIDP.31 In our series, patients with typical CIDP had a worse MRC score, higher disability and worse QoL than patients with DADS, confirming that DADS is clinically less disabling than typical CIDP. Patients with LSS had less severe disability than patients with typical CIDP by I-RODS but not by INCAT. This discrepancy may reflect the wider range of item difficulties in the I-RODS than INCAT scale. On the other hand, we did not find differences in the degree of motor impairment, disability and QoL between patients with purely motor CIDP and patients with typical CIDP. Similarly, disability and QoL where similar in sensory and typical CIDP.

    There are some differences in the reported response to therapy in patients with atypical CIDP. A better response to IVIg than to steroids was reported in some series of patients with LSS,4 5 7 16 17 while others6 reported a reduced response to IVIg but not to steroids compared with typical CIDP. Patients with DADS were reported to have a less satisfactory response to treatment than patients with typical CIDP.18 19 In our series, patients with LSS and DADS had a less frequent response to therapy (p=0.0433; p=0.0050) and to IVIg (p=0.0086; p=0.0178) compared with patients with typical CIDP, and the same occurred for patients with purely sensory DADS, while there were no differences in the response to steroids.

    Several reports suggest that purely motor CIDP may not respond to or even worsen after corticosteroids.20 21 Based on these findings, the EFNS/PNS Guidelines recommend the use of IVIg for motor CIDP.1 However, response to steroids has been reported in other studies, with figures up to 20% of treated patients.13 We found that 43% of our patients with purely motor CIDP improved after steroids. All these patients had however a concomitant sensory electrophysiological, although not clinical, impairment. This finding suggest that the diagnosis of motor CIDP and the decision to avoid steroids should be probably restricted to patients without any clinical and electrophysiological involvement of sensory nerves. Only few data are available on the response to therapies in patients with purely sensory CIDP with a usually favourable response to both IVIg and steroids.13 22–24 We also found a similar response rate in comparison witho patients with typical CIDP.

    In conclusion, our study on a large population of patients classified according to a uniform set of diagnostic criteria confirms that the proportion of patients with atypical CIDP varies according to the duration of the disease and that response to therapy is different in some of these forms (DADS and LSS). The persistence of an atypical presentation in a consistent proportion of patients for several years together with the different response to therapy in some of these forms suggests that some difference in the pathogenic mechanisms may underlie some of these variants. An extensive immunological study to investigate the presence of different anti-nerve antibodies is in progress in this cohort of patients to verify whether the different clinical presentations and response to therapy might be associated with specific immunological abnormalities as was recently reported for patients with antibodies to neurofascin 155, contactin 1 and other nodal–paranodal proteins.32 33 It would be also advisable to verify these data in a prospective study on a large series of newly diagnosed patients in whom a definite set of diagnostic criteria and of diagnostic investigations and assessment would be applied and subsequently periodically verified.34

    Supplementary data

    References

    View Abstract

    Footnotes

    • Collaborators Pietro Emiliano Doneddu, Giuseppe Liberatore, Francesca Gallia and Eduardo Nobile-Orazio from the Department of Medical Biotechnology and Translational Medicine, Neuromuscular and Neuroimmunology Service, Humanitas Clinical and Research Institute, Milan University, Rozzano, Milan, Italy; Erdita Peci and Dario Cocito from the Department of Neuroscience, University of Turin, Turin, Italy; Daniele Velardo, Stefano Tronci and Raffaella Fazio from the Division of Neuroscience, Department of Neurology, Institute of Experimental Neurology (INSPE), San Raffaele Scientific Institute, Milan, Italy; Fiore Manganelli and Lucio Santoro from the Department of Neuroscience, Reproductive Sciences and Odontostomatology, University of Naples ‘Federico II’, Naples, Italy; Marta Ruiz and Chiara Briani from the Neurology Unit, Department of Neuroscience, University of Padua, Padua, Italy; Stefano Cotti Piccinelli, Alice Todeschini and Massimiliano Filosto from the Center for Neuromuscular Diseases and Neuropathies, Unit of Neurology ASST ‘Spedali Civili’, University of Brescia, Brescia, Italy; Alessandro Beronio and Luana Benedetti from the Neurology Unit, Sant’Andrea Hospital, La Spezia, Italy; Antonio Toscano, Luca Gentile and Anna Mazzeo from the Department of Clinical and Experimental Medicine, Unit of Neurology, University of Messina, Messina, Italy; Giorgia Mataluni and Girolama Alessandra Marfia from the Disimmune Neuropathies Unit, Department of Systems Medicine, Tor Vergata University of Rome, Rome, Italy; Laura Piccolo, Ilaria Callegari and Andrea Cortese from the IRCCS Foundation C. Mondino National Neurological Institute, Pavia, Italy; Giuseppe Cosentino and Brigida Fierro from the Department of Experimental BioMedicine and Clinical Neurosciences (BioNeC), University of Palermo, Palermo, Italy; Verrengia Elena Pinuccia and Stefano Jann from the Department of Neuroscience, Niguarda Ca’ Granda Hospital, Milan, Italy; Elisa Bianchi and Ettore Beghi from the Laboratorio di Malattie Neurologiche, IRCCS-Istituto Mario Negri, Milan, Italy; Angelo Maurizio Clerici from the Neurology Unit, Circolo and Macchi Foundation Hospital, Insubria University, DBSV, Varese, Italy; Federica Scrascia and Marinella Carpo from the ASST Bergamo Ovest-Ospedale Treviglio, Treviglio, Italy; Martina Garnero and Angelo Schenone from the Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa and IRCCS AOU San Martino-IST, Genoa, Italy; Marco Luigetti and Mario Sabatelli from the Department of Neurology, Catholic University of Sacred Heart, Rome, Italy; Patrizia Dacci and Giuseppe Lauria from the Unit of Neuroalgology, IRCCS Foundation ‘Carlo Besta’ Neurological Institute, Milan, Italy; Luca Leonardi and Giovanni Antonini from the Unit of Neuromuscular Diseases, Department of Neurology Mental Health and Sensory Organs (NESMOS), Faculty of Medicine and Psychology, ‘Sapienza’ University of Rome, Sant’Andrea Hospital, Rome, Italy; Tiziana Rosso from the Azienda UL.SS. 8 Asolo, Castelfranco Veneto, Italy; Erika Schirinzi and Gabriele Siciliano from the Neurology Unit, Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy; Claudia Balducci and Guido Cavaletti from the School of Medicine and Surgery and Experimental Neurology Unit, University of Milano-Bicocca, Monza, Italy.

    • Contributors PED contributed to the conception of the research project, reviewed and commented on the statistical analysis, wrote the first draft of the report and reviewed the report. DC, FM, RF, CB, MF, LB, AM, GAM, AC, BF, SJ, AMC, MC, AS, ML, GL, GA, TR, GS, GC, GL, LS, EP, ST, MR, SCT, AT, GM, LP, GC and MS contributed to the conception, organisation and execution of the research project, and reviewed and commented on the statistical analysis and the report. EB designed and executed the statistical analysis, contributed to the conception, organisation and execution of the research project, and reviewed and commented on the statistical analysis and the report. EN-O conceived, organised and designed the study, reviewed and commented on the statistical analysis, wrote the first draft of the report and reviewed the report.

    • Funding The study was supported by a grant from Regione Lombardia, Italy, for patients from this region and subsequently extended to other Italian Centres. The study was also supported by unrestricted grants from Kedrion Biopharma (Italy), CSL Behring (Italy), Humanitas Clinical and Research Institute (Milan, Italy), and GBS-CIDP Foundation International (USA).

    • Disclaimer The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

    • Competing interests EN-O reports personal fees for Advisory or Scientific Board from Kedrion, Italy, Baxter, Italy, Novartis, Switzerland, CSL Behring, Italy, Astellas, the Netherlands, outside the submitted work and travel grants to attend scientific meeting from Baxter, Grifols, Kedrion and Novartis, Italy. PED has received travel grants to attend scientific meetings from CSL Behring and Kedrion. GLib has received travel grants to attend scientific meetings from CSL Behring and Kedrion. DC has received honoraria for lecturing from Shire, CSL Behring and Kedrion and travel grants to attend scientific meeting from Shire, Kedrion and CSL Behring. EP has received travel grants to attend scientific meetings from CSL Behring. RF has served on scientific advisory boards for CSL Behring and has received travel grants from Kedrion and CSL Behring to attend scientific meeting. MC has received travel grants to attend scientific meetings from Kedrion. AM has received travel grants from Kedrion and CSL Behring to attend scientific meeting. CB has served on scientific advisory boards for Pfizer and has received travel grants from Kedrion and CSL Behring to attend scientific meeting. GC has received travel grants to attend scientific meetings from CSL Behring and Kedrion. BF has received travel grants to attend scientific meetings from CSL Behring and liberal contribution from CSL Behring for the neuromuscular diseases centre, outside the submitted work. EB reports grants from UCB-Pharma, grants from Shire, grants from EISAI, personal fees from Viropharma, grants from Italian Ministry of Health, grants from Fondazione Borgonovo, grants from Associazione IDIC 15, grants from European Union, outside the submitted work. AC has received travel grants to attend scientific meetings from Kedrion. ML has received honoraria for scientific board from Pfeizer and Alnylam and travel grants from Grifols and Kedrion to attend scientific meeting. LS reports personal fees for scientific events from CSL Behring and has received travel grants to attend scientific meetings from CSL Behring and Kedrion. FM reports personal fees for scientific events from CSL Behring and has received travel grants to attend scientific meetings from CSL Behring and Kedrion. GC has received honoraria for lecturing and travel grants to attend scientific meetings from Kedrion. MF has served on scientific advisory boards for CSL Behring and has received travel grants from Kedrion, Baxter and CSL Behring to attend scientific meeting. SJ has received research grants from Grifols, outside this work, and travel grants from Grifols and Kedrion.

    • Patient consent Obtained.

    • Ethics approval The study was approved by the Ethical Committee of each participating Centre.

    • 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