Chronic infl ammatory demyelinating polyneuropathy and diabetes – a challenge for neurologists

Summary

Chronic inflammatory demyelinating polyneuropathy (CIDP) is an acquired, immune-mediated neuropathy caused by inflammation of peripheral nerves and nerve roots. It is the most common chronic autoimmune polyneuropathy, which is still considered underdiagnosed. If it remains untreated or improperly treated for a long time, it can lead to severe disability with impairment of the patient‘s fine motor skills, walking, and general mobility. CIDP may be associated with several diseases such as diabetes mellitus, monoclonal gammopathy, HIV infection, malignancies, or several systemic diseases. Recently, there have been several references that the prevalence of CIDP tends to be higher in diabetics, especially in older patients. Diagnosing CIDP in a patient with diabetes is challenging, because superimposed axonal damage in possible diabetic neuropathy can obscure typical demyelinating electrophysiological findings. On the other hand, diabetic polyneuropathy can cause elevated protein in cerebrospinal fluid. There are still many controversies in explaining the association of these two diseases. We still do not have an adequate diagnostic tool to clearly define CIDP in diabetic patients. The identifying a potential biomarkers of CIDP in diabetic patients is a challenge for neurologists, as CIDP is a treatable disease.

Keywords:

diagnostic criteria – Electromyography – diabetes mellitus – biomarkers – treatment – chronic infl ammatory demyelinating polyneuropathy – diabetic polyneuropathy

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

 

Introduction

Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is an acquired, immune-mediated polyneuropathy caused by inflammation of nerve roots and peripheral nerves and characterized by segmental demyelination and remyelination [1]. It is considered an underdiagnosed disease that can lead to severe incapacitation, yet is well treatable [2,3]. The prevalence of CIDP is estimated to be 0.67-10.3/100,000 -⁠ depending on the methodologies used, the incidence is less than 1.0/100,000/year [4,5].

In 2021, a joint expert working group of the European Academy of Neurology (EAN) and the Peripheral Nerve Society (PNS) [6] modified the previously valid EFNS/PNS criteria for CIDP [7]. According to this revision, a distinction is made between typical CIDP (accounting for more than half of the cases) and variant CIDP (formerly atypical CIDP). Typical CIDP is clinically characterized by a relapsing-remitting or progressive course (at least 8 weeks) with the development of symmetrical weakness of proximal and distal muscles, loss of sensation and reduced or absent tendon reflexes. Variants of CIDP share common features of demyelination and response to immune therapy with typical CIDP. We include the distal variant, the multifocal variant, and the focal, motor, and sensory variants of CIDP [6,8]. The basis of the diagnosis of CIDP is the fulfilment of a combination of clinical, electrodiagnostic and supportive criteria. In Table 1, we list the currently valid clinical criteria for CIDP and in Table 2, we summarize the electrodiagnostic criteria for typical CIDP [6]. The recognition of CIDP is challenging in some cases, mainly due to the heterogeneous nature of the disease [9-11]. Despite tremendous progress, there is no definitive and reliable diagnostic marker for CIDP to date. Several diseases can be associated with CIDP such as diabetes mellitus (DM), monoclonal gammopathy of undetermined significance (MGUS), IgM monoclonal gammopathy without anti-MAG (myelin-associated glycoprotein) antibodies, HIV infection, chronic active hepatitis, or systemic lupus erythematosus [1,3,12-14].

In clinical practice, we increasingly encounter the situation when CIDP is diagnosed in a diabetic patient. The clinical and electrodiagnostic manifestations of CIDP and diabetic neuropathy are similar and may overlap. This thesis discusses key aspects of the diagnosis and management of CIDP in patients with diabetes.

 

Diabetes mellitus and diabetic neuropathy

Diabetic neuropathy (DN) is the most common chronic organ complication of DM. Its incidence depends on the duration of DM and the long-term level of metabolic control [15]. Due to the late onset of type 2 DM, early symptoms of DN are often not recognized [16]. DN has several subtypes, which have different pathophysiology and distinct clinical picture. Among several classification systems, I report the division of DN according to P. K. Thomas et al. in 1997, which is still widely accepted today [17]. The most common form of DN (approximately 80% of cases) is diabetic symmetric predominantly sensory polyneuropathy (DSPN), in which the damage is nerve fiber length dependent (length dependent) [17,18]. However, there is a whole spectrum of less common symmetric, focal, and mixed forms of DN (Table 3) [17]. Another form of neuropathy that can occur in a diabetic patient is CIDP.

 

CIDP and diabetes mellitus

We know from several studies and from clinical practice that CIDP and DM rarely occur together. The possible relationship between DM and CIDP and its clinical significance have long been a controversial topic. The diagnosis of CIDP in a patient with DM is challenging because on the one hand superimposed axonal damage can mask the electrophysiological findings of CIDP, on the other hand DN can cause elevated protein levels in the liquor [19]. Several studies have shown that the incidence of CIDP is higher in diabetic patients than in the general population [19,20]. In 2016, Canadian and American authors [21], conducted an extensive literature search of the PharMetrics Plus™ database (years 2009-2013). The prevalence of CIDP among non-diabetic patients was 6/100,000, in contrast, the prevalence of CIDP in the diabetic population was 9 times higher (54/100,000). In the work of Doneddu et al. [22], DM was also more prevalent (14%) compared to the general European population. Patients with DM had higher disability scores, poorer quality of life and worse response to treatment than CIDP patients without DM.

In some other papers, the association between CIDP and DM has not been confirmed [23]. Most authors report that CIDP is more common in patients with type 2 DM [24], while others have not confirmed this [20]. In a study by Dunnigan et al. [24], patients with CIDP and concomitant DM (DM-CIDP) showed more severe axonopathy, probably due to the associated diabetic axonopathy. In contrast to DPN patients (n = 56), DM-CIDP patients (n = 67) had more pronounced motor nerve conduction slowing, were older overall (65.1 ± 13.7 years vs. 55 ± 16 years), had more severe neuropathy, shorter duration of DM, and better glycemic control compared with diabetic patients without CIDP.

There is considerable uncertainty regarding possible common pathophysiological mechanisms between CIDP and DN. Most respected authors are inclined to the view that DM is unlikely to be a major risk covariate for CIDP. It is still unclear whether this is a causal relationship or a coincidental phenomenon [25,26].

 

How to distinguish DM-CIDP from "pure" DSPN?

In clinical practice, we encounter a situation when we have a diabetic patient with slowly progressive proximal and distal muscle weakness and present demyelinating findings on EMG. We are considering possible CIDP, and the liquor examination shows a proteinocytological dissociation. However, we know that elevation of proteins in the liquor is also common in DSPN, and even more common in proximal diabetic amyotrophy. In this situation, how do we differentiate DM-CIDP from DSPN ?

DSPN is characterized by early involvement of the thin fibers, leading to pain and dysesthesias; only later are the thick fibers affected, resulting in sensations of numbness and loss of prosthetic sensitivity. On objective examination, we observe a distal deterioration of sensitivity to touch and pinprick with a sharp object. Typically, sensory impairment predominates over motor impairment, proprioception is impaired with subsequent ataxia, symptoms are symmetrical and progression is slow. Significant motor impairment is present only in advanced stages of DSPN. Autonomic nerve involvement is common [19,21,27-29].

In contrast, in CIDP, the autoimmune process is directed primarily against myelin, the main carrier of which is the thick nerve fibres. Thin fibres are not significantly affected. The clinical picture of CIDP is based on this fact. Rapid progression of motor weakness is typical, proximal and distal weakness is present simultaneously, and tendon reflexes are often completely absent. In the sensory area, deep sensitivity (hemi-sensitivity, motion sensitivity, vibratory sensitivity) is predominant, resulting in a relatively early onset of sensory ataxia. Acral paresthesias are common; in contrast, neuropathic pain and involvement of autonomic nerve fibres is very rare in DSPN [19,21,27,28].

The question must be asked: "Can electrophysiological testing help us differentiate DM-CIDP from pure DSPN?" It has been shown in several studies that the degree of demyelinating manifestations (prolonged distal motor latencies, prolonged/absent F wave, temporal dispersion of the summatory muscle action potential (CMAP), etc.) was more pronounced in the DM-CIDP group than in pure DSPN. Patients in the DM-CIDP group were older, had a shorter duration of DM, and better glycemic control. We can say that if a given diabetic patient meets the EMG demyelinating criteria for CIDP, it is highly likely to be DM-CIDP and not DSPN. Another distinguishing criterion is increased temporal dispersion of the distal CMAP, which is very rare in axonal neuropathies [19,21,28,30,31].

In practice, we may encounter the need for differential diagnosis of DM-CIDP and proximal diabetic amyotrophy (PDA). Here, the asymmetry of findings in PDA (asymmetric weakness and hypotrophy of the thigh, absence of reflexes in one limb) is important. There is also bilateral PDA, but it is always asymmetric and tends to be very painful [28,32-34]. Examination of the liquor is not diagnostic as both DM-CIDP and PDA tend to have hyperproteinorachia [28].

Another tool to differentiate between DM-CIDP and DSPN is the use of neuroimaging scans. Magnetic resonance neurography (MRN) of the brachial and lumbosacral plexus in CIDP patients can show hypertrophy, hyperintensities, and postcontrast enhancement of roots and nerves. The differentiation of CIDP from proximal diabetic amyotrophy by MR techniques is the subject of further debate [28,35].

In recent years, there has been a growing body of work on the use of high-resolution US to refine the diagnosis of several neuropathies [36,37]. In general, demyelinating neuropathies are more commonly associated with nerve enlargement than axonal neuropathies because ultrasound has the ability to detect nerve inflammation or swelling. Nerve enlargement (thickening) in CIDP is often regionally or focally limited and has a different pattern of echointensity. US results of peripheral nerves in DSPN tend to be heterogeneous and nerve enlargement tends to be milder compared to CIDP [38-40] (Figure 1) [41]. Nerve (n. suralis) biopsy is currently extremely rarely performed in the diagnosis of CIDP and is less sensitive and specific than electrodiagnostics. Some authors suggest the use of corneal confocal microscopy in the differential diagnosis of DM-CIDP and DSPN [42].

A more accurate identification of DM-CIDP versus DSPN has been attempted by Israeli authors [43], in their algorithm they propose to use a combination of clinical, EMG and laboratory parameters. They assigned a positive (+) value to some parameters supporting CIDP and questioning the diagnosis of DPN (Table 4); and they assigned a (-) value to parameters questioning CIDP and supporting DPN (Table 5). Although this tool has been validated in a small number of patients (n = 57), it correlates well with The American Academy of Neurology (AAN) criteria [44] as well as the European Federation of Neurological Societies (EFNS) criteria for CIDP [7].

According to the total score, the authors defined 4 levels of certainty for the diagnosis of CIDP in a diabetic patient [43]:

a) ≥ 11 points -⁠ diagnosis certain;

b) 5-10 points -⁠ diagnosis probable;

c) 2-4 points -⁠ diagnosis possible;

d) 0-1 point -⁠ diagnosis unlikely.

 

Currently, there are no specific diagnostic criteria for patients with DM-CIDP. Diagnosis is based on the evaluation of clinical, electrophysiological and neuroimaging examinations in combination with dg. criteria for CIDP. In simplified terms, when a patient subacutely develops painless, progressive, distal and proximal limb weakness, with an anamnestic shorter duration of DM, a diagnosis of CIDP should be considered [6,7,14,28,33].

When CIDP is highly suspected, especially in the setting of disabling neuropathy, a treatment trial is recommended -⁠ intravenous immunoglobulin (IVIg) or corticosteroids (CS) [6,29,43]. Patients with DM-CIDP usually respond well to immunotherapy and improve clinically.

 

Treatment of CIDP patients with diabetes

Currently, the decision on how to treat CIDP in diabetic patients is guided by the treatment of CIDP without DM [26]. First choice treatments include IVIg and CS. When these are ineffective, plasmapheresis (PF) or the deployment of some immunomodulatory drugs are considered [6,28]. There are several studies that confirm that DM-CIDP patients, as well as CIDP patients without DM, respond positively to immunologic therapy [20,45-47]. There is a tendency for physicians to deploy IVIg as the first choice in DM-CIDP patients because of the concern about possible decompensation of DM with steroid therapy. However, this tendency is not clearly supported by evidence-based data. Also according to the Toronto data, IVIg is more frequently deployed as the first choice in both CIDP and DM-CIDP patients, second only to CS [21].

A 2015 paper by Canadian authors [47] compared the therapeutic response in a total of 99 CIDP patients, 34 had both CIDP and DM, 65 had CIDP without DM. Both groups had a similar degree of therapeutic response. They found that the "stronger responders" were characterized by meeting more AAN and EFNS/PNS criteria overall and having a higher number of EMG demyelinating manifestations.

When choosing immunomodulatory therapy, a personalized approach should be followed and the patient's comorbidities (arterial hypertension, osteoporosis, cardiac disease), age, short -⁠ and long-term efficacy of the drugs, their side effects, ease of administration, and cost should be considered. IVIg should be preferred in more severe forms of the disease where we need to achieve a rapid effect, in patients with comorbidities, in elderly patients, and in cases of contraindications to the administration of CS. The disadvantages of IVIg may be the high cost and possible side effects (flu-like, cephalea, risk of thromboembolism) [48-50]. In younger DM-CIDP patients without comorbidities, with a milder course, oral prednisone may be an option. KS should be avoided in poor glycaemic control, infection, hyperadrenocorticism and peptic ulcer. Adverse effects of therapy (osteoporosis, obesity, cataract, ulcer disease) should be monitored and addressed. Larger clinical trial data are lacking on the long-term use of CS in DM-CIDP patients [51,52].

Second choice procedures are plasmapheresis (PF) and the use of immunosuppressive and immunomodulatory agents. PF is a challenging procedure (instrumentation, venous access, risks of infection) that has short-term efficacy and therefore requires repeated application [50,51,53]. Regarding the efficacy of other drugs (azathioprine, methotrexate, cyclosporine, cyclophosphamide, rituximab) in DM-CIDP, we lack any data from clinical trials. For maintenance treatment of DM-CIDP we have IVIg, subcutaneous administration of immunoglobulin (SCIg) or (prednisone), doses of drugs should be strictly personalized.

 

Conclusion

Chronic inflammatory demyelinating polyneuropathy is the best treatable neuropathy that can occur in a diabetic. What the exact relationship between diabetes and CIDP is remains controversial. Several respected authors believe that DM may be a risk factor for CIDP, but we lack data from randomized trials.

Diagnosing CIDP in a diabetic patient due to concomitant axonal damage is often difficult and ambiguous. DM-CIDP should be thought of when anamnestic muscle weakness is brief, proximal weakness is present, significant upper limb involvement is present, and it is type 2 DM (often well compensated) with shorter duration. The electrophysiological criteria for CIDP are mandatory. Patients with coexisting CIDP and DM have similar symptoms to those with 'pure' CIDP, but have more severe balance impairment and more pronounced axonal damage.

When CIDP is highly suspected, a therapeutic test -⁠ IVIg or KS -⁠ should be initiated; patients respond well to immunomodulatory therapy, but improvement may be less pronounced and slower. We currently lack reliable biomarkers to accurately differentiate DM-CIDP from DPN. The use of peripheral nerve ultrasonography or MR neurography may be a useful tool in diagnosis. Another aid is the use of the algorithm by Lotan et al. [43].

 

Table 1 Clinical criteria for CIDP -⁠ adapted from [8].

Typical CIDP

All criteria met:

Progressive or relapsing, symmetrical, proximal and distal muscle weakness on DK and HK, sensory involvement in at least two limbs

development of symptoms lasts at least 8 weeks

absence or reduction of tendon reflexes in all limbs

CIDP variants

One criterion for the variant is met, but others as for typical CIDP. Tendon reflexes may be normal in unaffected limbs:

Distal CIDP: Distal sensory impairment, distal muscle weakness with predominance on DK

multifocal CIDP: multifocal pattern of muscle weakness and impaired sensation, usually asymmetrical, with predominance on the DK, in at least two limbs

focal CIDP:  sensory impairment and muscle weakness in only one limb

motor CIDP: muscle weakness without sensory impairment

Sensory CIDP: sensory impairment without muscle weakness

CIDP: chronic inflammatory demyelinating polyneuropathy; DK: lower limbs; HK: upper limbs

 

 

Table 2. Electrodiagnostic criteria for typical CIDP -⁠ criteria for motor conduction studies -⁠ adapted from [8].

1. Strongly promoting demyelination (criterion for dg. CIDP):

At least 1 of the following points a-g is met:

(a) DML prolonged ≥ 50% above the upper limit in at least 2 nerves (excluding carpal tunnel syndrome)

(b) reduced motor RV ≥ 30% below the lower limit in at least 2 nerves

(c) prolonged F-wave latency ≥ 20% above upper limit in at least 2 nerves (≥ 50% ampl distal CMAP < 80% of lower limit

(d) absence of F-wave in at least 2 nerves (if ampl of distal CMAP ≥ 20% of lower limit) + at least 1 parameter of demyelination in at least 1 other nerve)

(e) motor conduction block: ≥ 30% reduction in ampl CMAP on proximal stimulation (except n.tibialis) and ampl distal CMAP ≥ 20% of lower limit in 2 nerves; or in 1 nerve + at least 1 parameter of demyelination except absence of F-wave

(f) Abnormal temporal dispersion: >30% prolongation of CMAP duration on proximal stimulation versus distal stimulation in at least 2 nerves (at least 100% prolongation in n.tibialis)

(g) prolongation of CMAP duration on distal stimulation in at least 1 nerve + at least 1 additional parameter of demyelination in at least 1 additional nerve. At 20 Hz low-pass filter: n. medianus > 7.4 ms, n. ulnaris > 7.8 ms, n. peroneus > 8.1 ms, n. tibialis > 8.0 ms

2. Poorly supporting demyelination: (criterion for dg. possible CIDP):

Same as in point 1, but only in 1 nerve.

CIDP, chronic inflammatory demyelinating polyneuropathy; CMAP, summatory muscle action potential; DML, distal motor latency; RV, conduction velocity 

 

 

 

 

Table 3. Classification of diabetic neuropathies according to F. K. Thomas [16] -⁠ modified.

Hyperglycaemic neuropathy

Symmetrical polyneuropathies distal, predominantly sensitive polyneuropathy Autonomic neuropathy acute painful diabetic neuropathy pseudotabic neuropathy

Focal and multifocal neuropathies cranial neuropathies thoracoabdominal neuropathy focal limbic neuropathies proximal diabetic amyotrophy

Mixed forms

 

 

Table 4. Diagnostic criteria for CIDP in diabetes according to Lotan et al [40] -⁠ modified.

Parameters supporting CIDP.

	Parameters	Score
1.	Clinical
	Progressive/relapsing motor weakness over 2-6 months	+3
	Significant proximal motor weakness	+3
	Significant HK involvement (either symmetrical or asymmetrical)	+3
	impairment of deep sensation > superficial sensation (if sensory manifestations predominate)	+2
	recent onset of DM, relatively well compensated	+1
2.	Electrophysiological
	Prolonged distal motor latency ≥ 50% -⁠ in at least 2 nerves	+1
	slow motor conduction velocity ≤ 30% -⁠ in at least 2 nerves	+1
	prolonged F wave latency ≥ 20% -⁠ at least in 2 nerves	+1
	partial motor conduction block (at least 50% reduction in proximal versus distal CMAP; or abnormal CMAP temporal dispersion (>30% increase in duration between proximal and distal CMAP)	+3
	distal CMAP duration ≥ 9 ms in at least 1 nerve + at least 1 additional typical "demyelinating" parameter in at least 1 additional nerve	+3
3.	Supplementary
	Proteinorachia ≥ 2 times the upper limit	+2
	Associated serological biomarkers suggestive of other dysimmune inflammatory diseases (Ery sedimentation, CRP, ANA, dsDNA, anti-Ro, anti-La, RF, paraprotein, etc.)	+1

ANA, antinuclear antibodies; CMAP, summated muscle action potential; CRP, C-reactive protein; DM, diabetes mellitus; dsDNA, double-stranded deoxyribonucleic acid; Ery, erythrocytes; HK, upper limbs; RF, rheumatoid factor

 

 

 

 

 

 

 

Table 5. Diagnostic criteria for CIDP in diabetes according to Lotan et al. [40] -⁠ modified

Parameters questioning CIDP.

	Parameters	Score
1.	Clinical
	Slowly progressive course (> 1 year)	-2
	Predominantly sensitive symptoms suggesting thin fiber involvement	-2
	involvement of cranial nerves (except n. facialis) and/or involvement of autonomic nerves	-3
	preserved deep tendon reflexes	-3
2.	Electrophysiological
	CMAP amplitude reduction in disproportion to motor conduction velocity	-3
3.	Histology
	Axonal degeneration without evidence of demyelination	-2
	Evidence of other causes of neuropathies that may resemble CIDP (vasculitis, sarcoidosis, amyloidosis)	-3

CIDP -⁠ chronic inflammatory demyelinating polyneuropathy; CMAP -⁠ summatory muscle action potential