Abstract
Objective: To characterize an acquired, symmetric, demyelinating neuropathic variant with distal sensory or sensorimotor features.
Background: Classic chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) patients have prominent proximal and distal weakness. However, chronic demyelinating neuropathies may present with different phenotypes. An approach that distinguishes these disorders primarily according to the pattern of weakness may be useful to the clinician.
Methods: A total of 53 patients with acquired symmetric demyelinating polyneuropathies were classified primarily according to the pattern of the neuropathy and secondarily according to the presence and type of monoclonal protein (M-protein) in this retrospective review. The authors distinguished between patients with distal sensory or sensorimotor involvement, designated as distal acquired demyelinating symmetric (DADS) neuropathy, from those with proximal and distal weakness, who were designated as CIDP.
Results: M-proteins were present in 22% of patients with CIDP. There were no features that distinguished clearly between CIDP patients with or without an M-protein, and nearly all of these patients responded to immunomodulating therapy. In contrast, nearly two-thirds of the patients with DADS neuropathy had immunoglobulin M (IgM) kappa monoclonal gammopathies, and this specific combination predicted a poor response to immunomodulating therapy. Antimyelin-associated glycoprotein (anti-MAG) antibodies were present in 67% of these patients.
Conclusion: Distinguishing acquired demyelinating neuropathies by phenotype can often predict the presence of IgM kappa M-proteins, anti-MAG antibodies, and responses to immunomodulating therapy.
The clinical hallmark of acute and chronic inflammatory demyelinating polyradiculoneuropathies (CIDPs) is the presence of proximal as well as distal weakness.1,2 Dyck et al.2 emphasized a pattern of weakness “tending to affect distal and proximal limbs” as an important feature favoring the diagnosis of CIDP. Patients with both proximal and distal motor involvement have a high probability of responding to corticosteroids, irrespective of the presence or absence of a serum paraprotein.3 This observation has led to the suggestion that proximal weakness be a mandatory clinical criterion for diagnosis.3
Another well-documented group of patients with symmetric, acquired, demyelinating polyneuropathies present with predominantly sensory involvement. When weakness is present, it is limited to distal muscle groups in a length-dependent fashion. Despite the prominence of sensory symptoms, motor nerve conduction studies reflect a demyelinating process. Prior reports have categorized these patients primarily on the basis of the presence of an immunoglobulin M (IgM) monoclonal protein (M-protein) or antibodies reactive to myelin-associated glycoprotein (MAG).4-6 Although a few studies describe benefits from intensive immunomodulating therapy (IMT),7-10 other reports suggest that, in contrast to classic CIDP, these patients often fail to respond to standard regimens of oral prednisone, IV immunoglobulin (IVIg), or plasmapheresis.4-6,11,12 We have adapted a term for this phenotype: distal acquired demyelinating symmetric (DADS) neuropathy.
We performed a retrospective analysis of our experience with symmetric acquired demyelinating neuropathies, both CIDP and DADS, to determine how distinct these phenotypes are with regard to the presence of M-proteins, classes of these proteins, electrodiagnostic findings, and response to IMT. We discuss the utility of this approach, which distinguishes polyneuropathies primarily by clinical and electrodiagnostic features, as opposed to the more common distinction that emphasizes the presence or absence of serum M-proteins or antibodies directed against peripheral nerves.
Methods.
Patient selection.
Clinical and electrophysiologic data were reviewed retrospectively for 53 consecutive patients with symmetric acquired neuropathies who fulfilled electrodiagnostic criteria for a demyelinating polyneuropathy.13 Forty-five patients were seen at The University of Texas Southwestern Medical Center between January 1, 1995, and June 1, 1999, and eight patients were seen at Stanford University Medical Center between January 1, 1998, and June 1, 1999. Patients with asymmetric acquired demyelinating neuropathies, including multifocal motor neuropathy13 or multifocal acquired demyelinating sensory and motor neuropathy,14 were not included in this analysis. We excluded patients if our workup revealed a lymphoproliferative disorder or a secondary cause for the neuropathy.
The classification of phenotype was based on the neurologic examination held before treatment. Patients with a distal, symmetric, sensory or sensorimotor neuropathy sparing proximal limb, neck, and facial muscles were designated as having DADS neuropathy. This length-dependent, primarily sensory phenotype affected the distal lower extremities and, in more advanced cases, the hands. In contrast, patients with CIDP presented with proximal and distal weakness involving all four limbs and, at times, affecting neck or facial muscles. Motor symptoms tended to dominate the clinical picture whereas sensory loss occurred in a stocking or stocking–glove distribution.
Clinical measures.
Manual muscle testing was performed in 34 muscle groups using the modified Medical Research Council (MRC) 10-point grading system at the time of initial presentation and during all subsequent visits. We also performed a functional assessment using a modified Rankin scale: 0, asymptomatic; 1, nondisabling symptoms; 2, minor disability; 3, moderate disability that interferes markedly with lifestyle or limits independent existence; and 4, moderately severe disabling symptoms that prevent independent existence (although the patient does not need constant attention).
Electrodiagnostic studies.
We performed standard nerve conduction studies using supramaximal stimulation and recording from surface electrodes as described previously.13 We evaluated median, ulnar, posterior tibial, and peroneal motor nerve conduction studies and median, ulnar, and sural sensory responses in each patient. The specific parameters of demyelination analyzed in this report included the motor conduction velocity, motor distal latency, and terminal latency index (TLI) using the formula
TLI = distal distance/(proximal conduction velocity × distal latency)
The TLI was considered abnormal if it was more than 2 SDs below our mean value for normal control subjects.15
Laboratory studies.
All patients underwent a complete blood count, liver function studies, routine chemistries, sedimentation rate, vitamin B12, antinuclear antibody, and thyroid function tests. HIV and hepatitis panels were ordered in at-risk individuals, and other testing was ordered at the discretion of the treating physician.
Serum protein electrophoresis and immunofixation electrophoresis for IgM, IgG, IgA, kappa, and lambda were obtained in all 53 patients. Anti-MAG antibody titers were available in the majority patients with IgM M-proteins. These antibodies were not measured routinely in other patients. A commercial laboratory (Athena Diagnostic Laboratories, Worcester, MA) performed the anti-MAG antibody assay. According to information distributed by the laboratory, they prescreen sera for reactivity to sulfated galactosyl paragloboside using thin-layer chromatography. Titers of anti-MAG antibodies are then determined for positive sera by ELISA. Positive ELISAs are verified by Western blot analysis for MAG.
We classified patients secondarily based on the presence or absence of an M-protein. CIDP patients were separated into CIDP-M (with an M-protein) and CIDP-I (idiopathic). Similarly, DADS patients were classified as DADS-I neuropathy and DADS-M neuropathy. DADS-M patients were also grouped according to the specific class of M-protein.
Treatment.
We treated patients with prednisone, IVIg, plasmapheresis, or oral Cytoxan. We used a similar approach for treating CIDP and DADS patients during the period of this study. Prednisone was the most commonly used first line of therapy, with an initial dosage of 1.0 to 1.5 mg/kg daily. The dosage was reduced when patients demonstrated improvement or was discontinued if there was no marked improvement after 2 to 3 months. IVIg was used as an initial mode of therapy or as a second-line agent for patients who failed to respond to prednisone. The initial dosage was 2 g/kg over 2 to 5 days in equal, divided doses. Repeat 0.4-g/kg treatments were given at regular intervals, usually monthly, after the initial course. Patients were assessed for improvement from 2 to 4 weeks after the completion of each course, and treatments were discontinued if there was no response after 3 months. Plasma exchange or oral Cytoxan were used as second- or third-line treatments.
End point measures and statistical analysis.
Clinical and electrodiagnostic measures, as well as responses to treatment of patients with DADS-I neuropathy, DADS-M neuropathy, and CIDP were compared. Categoric data were compared using the chi-square or Fisher’s exact test as appropriate. Means were compared using one-way analysis of variance. Post hoc adjustments were performed using Tukey’s honestly significant difference. We analyzed clinical features including duration of illness, age, and loss of sensation, and electrodiagnostic features including distal motor latency, motor nerve conduction velocity, compound motor action potential amplitude, and TLI.
With respect to treatment, patients were dichotomized into responders and nonresponders for each individual treatment, as well as for the overall course of therapy. A response was graded positive when a patient improved by at least 1 point on the Rankin scale or by one motor grade in any muscle group on MRC testing. We performed a separate analysis by questioning patients as to whether they believed they had any subjective sensory improvement with treatment. Nonresponders showed no motor improvement and reported no sensory improvement. Response to therapy was compared between each group as defined earlier.
Results.
Clinical characteristics.
Of the 53 patients in this study, 23 patients had CIDP and 30 had DADS (table 1). Within the CIDP group, five patients (22%) had M-proteins (four patients, IgG kappa; one patient, IgM lambda). The average age at onset for CIDP was 52 years (range, 22 to 81 years) with a mean disease duration of 18 months. Weakness was present in all 23 patients and sensory symptoms were present in all but one patient. There were no salient clinical or electrodiagnostic differences between patients with CIDP-M and CIDP-I for any of the parameters measured, such that we combined the two groups for additional analysis against DADS-I neuropathy and DADS-M neuropathy. We also looked specifically for any differences between patients with DADS-M neuropathy versus CIDP-M.
Comparison of clinical and laboratory characteristics
Thirty patients had DADS neuropathy. Serum paraproteins were detected in 20 of these patients (67%; 18 patients, IgM kappa; 2 patients, IgG kappa). For DADS-M neuropathy, the average age at onset was 62 years (range, 35 to 81 years) and the average duration of symptoms was 60 months (range, 3 to 180 months). Nineteen of 20 patients were men. Eight patients had purely sensory signs and 12 patients had weakness at the ankles. For DADS-I neuropathy the average age at onset was 47 years (range, 27 to 67 years) with an average duration of 44 months (range, 1 to 144 months). Five DADS-I neuropathy patients had purely sensory signs and five had ankle weakness.
There were differences in the overall frequency of M-proteins (p = 0.002) and the frequency of IgM paraproteins (p < 0.001) between the CIDP and DADS phenotypes. We did not detect any significant differences for IgG paraproteins. Our clinical data showed that compared with DADS-M neuropathy, both CIDP and DADS-I neuropathy patients were significantly younger and were less likely to be men. The duration of symptoms was also significantly less in CIDP patients compared with DADS-M neuropathy. The only significant difference between DADS-I neuropathy and CIDP was a shorter duration of symptoms in CIDP patients. Comparing CIDP-M to DADS-M, we found differences for gender (p = 0.01) and for treatment response (p = 0.04). Removing the two patients with DADS-M neuropathy who had IgG M-proteins, to focus only on the subset of patients with DADS-M neuropathy and IgM paraproteins, had no effect on our analysis.
Electrodiagnostic measures.
We could not demonstrate any significant differences between routine electrophysiologic parameters among DADS-I neuropathy, DADS-M neuropathy, and CIDP (table 2). The mean median nerve TLI for DADS-M neuropathy patients was 0.25 (range, 0.12 to 0.40). DADS-I neuropathy patients had a mean median nerve TLI of 0.30 (range, 0.07 to 0.62). CIDP-I patients had a mean median nerve TLI of 0.33 (range, 0.09 to 0.52). We also did not detect significant differences between CIDP-M and DADS-M neuropathy.
Comparison of electrodiagnostic findings
Anti-MAG reactivity.
Anti-MAG antibodies were present in 10 of 15 patients tested with DADS-M neuropathy and IgM M-proteins (67%). No clinical or electrodiagnostic parameter reached significance between the MAG-reactive and MAG-nonreactive patients in this group. Anti-MAG antibodies were also present in the one patient with CIDP-M who had an IgM lambda M-protein. They were also detected in one of two patients tested with DADS-I neuropathy.
Treatment.
Overall, 95% of CIDP patients responded to treatment (table 3). This included objective responses in all five patients with CIDP-M. Four of five patients with DADS-I neuropathy reported subjective improvement in sensation with treatment compared with 3 of 10 patients with DADS-M neuropathy. Motor examination demonstrated improvement in two of three patients with DADS-I neuropathy compared with none of the eight patients treated with DADS-M neuropathy who exhibited weakness. There were significant differences in motor and subjective sensory improvement between DADS-I neuropathy and DADS-M neuropathy, as well as between CIDP and DADS-M neuropathy.
Treatment response
Discussion.
We performed this retrospective study to demonstrate the usefulness of a clinical approach to chronic, acquired, symmetric demyelinating polyneuropathies. Our method recognizes two distinct clinical presentations: the length-dependent, primarily sensory phenotype of “DADS neuropathy” and the predominantly motor phenotype with proximal and distal weakness of classic CIDP. A second purpose was to raise questions regarding the appropriateness of antibody-based designations for peripheral neuropathies.
All patients in this study with an IgM kappa M-protein fell into the group with distal involvement or DADS-M neuropathy, supporting the notion that this represents a distinct clinical entity. These patients represented the majority of the distal demyelinating phenotype and were almost uniformly older men with slowly progressive neuropathies. On nerve conduction testing, prolonged distal latencies with small TLIs were near-constant features. The presence of DADS-M neuropathy strongly predicted nonresponsiveness to IMT. Two-thirds of these patients had anti-MAG antibodies and, as a result, the syndrome corresponds closely to the majority of reports under the heading of IgM paraprotein- or anti-MAG antibody-associated neuropathy.4-6,16,17 Histopathology and passive transfer studies in this population support a direct autoimmune role for IgM antibodies for patients with18-20 and without anti-MAG reactivity.5,18 In our series, there were no salient features that differentiated MAG-reactive from nonreactive DADS-M neuropathy patients.
Only one-third of patients with the distal phenotype did not have M-proteins by SPEP and IFE. We refer to this presentation as DADS-I neuropathy. Five of 10 patients had purely sensory clinical involvement and conform to what has been described in the literature as “sensory CIDP.” Cases reported under this heading show electrodiagnostic evidence of subclinical motor demyelination21,22 and can evolve into typical CIDP,21 although we did not witness motor progression in our series. The other five patients had sensory and motor involvement, and might be viewed as a length-dependent CIDP variant. However, it is not clear that the DADS-I syndrome represents a single clinical entity. For example, one of these patients had generalized slowing of conduction and had the smallest TLI in our series (0.07) but lacked anti-MAG antibodies, which have been linked closely to this electrodiagnostic pattern.15,23 At the other end of the spectrum, DADS-I was associated with normal distal latencies, and conduction slowing was present only in more proximal nerve segments. Another patient with DADS-I had high titers of anti-MAG antibodies without a detectable M-protein. Others have also reported distal demyelinating neuropathies with anti-MAG or antisulfatide antibodies that lack detectable M-proteins, who therefore fall into the DADS-I syndrome.24,25 These findings raise the possibility that a subset of patients with DADS-I neuropathy have a disorder related physiologically or immunologically to DADS-M neuropathy, perhaps with antibodies causing neuropathic signs and symptoms at lower serum concentrations.
Because our study was retrospective and uncontrolled, we are cautious about making strong statements regarding treatment responses. However, in contrast to distal neuropathies, our findings confirm that patients with proximal and distal weakness are likely to respond to standard IMT, regardless of the presence or absence of an M-protein.3 There were also differences between CIDP-M and DADS-M neuropathy in gender and treatment response, and between the CIDP and DADS phenotypes in the incidences of M-proteins and IgM paraproteins. These observations raise questions about the research criteria used by the ad hoc subcommittee for the diagnosis of CIDP, which lumps patients under the heading of “concurrent diseases with CIDP” based solely on the presence of a paraprotein, without regard for the phenotype.26 Our analysis suggests that patients with CIDP-M have more in common with idiopathic CIDP than with DADS-M neuropathy.
Our method of grouping patients by clinical examination differs from previous reports on this topic, which have differentiated patients primarily by antibody class. Had we focused on antibodies, the IgM cohort would have included 18 patients with DADS and 1 patient with CIDP, whereas our IgG cohort would have included 2 patients with DADS and 4 patients with CIDP. Patients with primary length-dependent axonopathies may also have M-proteins,27 and had we included axonopathies evaluated over the same period we would have added six more patients to both groups. Some studies using this “antibody approach” have revealed that polyneuropathies associated with IgM monoclonal gammopathy are clinically28,29 or electrodiagnostically30 distinct from those associated with IgG, whereas others have not been able to demonstrate clear differences.31 This methodology has been criticized because the resulting two syndromes are heterogeneous32 and the significance of individual phenotypes is diluted. In our series, the DADS neuropathic phenotype would dominate in the IgM cohort, but the majority of patients in both groups would have distal phenotypes. Grouping patients with distinct clinical syndromes primarily by antibodies would have greater justification only if there was clear evidence of pathogenic mechanisms that united these heterogeneous neuropathies.
What are the practical implications of our findings? If a patient presents with a slowly progressive, acquired, distal sensory or sensorimotor neuropathy, statistically this most likely represents an axonal process.33 However, if electrophysiology reveals evidence of demyelination, the patient falls into the category of DADS neuropathy. Laboratory testing will reveal that the majority of these patients have an IgM kappa M-protein, particularly if they are older men. It is this specific phenotype and paraprotein combination that predicts a poor response to standard IMT. Anti-MAG antibodies can be detected in the majority of DADS-M neuropathy patients, but those with and without anti-MAG antibodies appear to be clinically, electrically, and therapeutically indistinguishable.
Finally, a stepwise approach—in which a neuropathy is characterized first by the clinical examination, then by electrophysiology, then by the presence and type of M-protein, and only subsequently by the specific immune response—is also implicit in other antibody-mediated or antibody-associated peripheral nervous system disorders. Consider that MG, Lambert–Eaton syndrome, multifocal motor neuropathy, and Guillain–Barré syndrome with its variants are also delineated primarily on the basis of a characteristic clinical and electrodiagnostic presentation. For each of these, the elucidation of an autoantibody has increased the understanding of the pathophysiologic processes or has facilitated diagnosis. Despite this, autoantibodies have never been employed as the primary criteria for defining a clinical entity. Clearly, we use the term MG as opposed to “acetylcholine receptor antibody syndrome,” and we are willing to make the diagnosis when antibodies are not present. Classifying patients as DADS neuropathy with IgM kappa M-proteins, as opposed to using the less specific terms “anti-MAG neuropathy” or “IgM monoclonal gammopathy of undetermined significance neuropathy” refocuses the clinician on a homogeneous clinical and electrophysiologic syndrome.
Footnotes
-
The views expressed herein are those of the authors and do not reflect the official policy of the United States Air Force or the Department of Defense.
- Received June 17, 1999.
- Accepted September 24, 1999.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Hastening the Diagnosis of Amyotrophic Lateral Sclerosis
Dr. Brian Callaghan and Dr. Kellen Quigg