Does the chronic inflammatory demyelinating polyradiculoneuropathy due to secondary cause differ from primary?
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.86537
Source of Support: None, Conflict of Interest: None
Background: The clinical presentation, neurophysiological findings, and outcome may vary between primary and secondary chronic inflammatory demyelinating polyradiculopathy (CIDP). Objective: To compare clinical and electrodiagnostic features of primary and secondary CIDP. Setting: Tertiary care teaching referral hospital. Materials and Methods: The CIDP patients who were diagnosed as per European Federation of Neurological Societies/Peripheral Nerve Society criteria were included and subjected to detailed history and examinations. The clinical disability was graded on a 0-10 scale. Neurophysiology included motor and sensory nerve conductions and F wave studies of all four limbs. Based on investigations for underlying diseases, the patients were categorized into primary or secondary CIDP. Prednisolone was prescribed in all and azathioprine added in resistant cases. The secondary CIDP group received specific treatment in addition. The outcome was assessed at 3 months, 6 months, and last follow-up. Results: A total of 65 patients aged 17 to 72 years were included and 20 were females. Twenty-five patients had secondary CIDP and include diabetes mellitus (16), POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes) (4), monoclonal gammopathy of undetermined significance (2), myeloma (1), lymphoma (1), and malignancy (1). The secondary CIDP patients were older (48.35 vs 41.0 years), had less relapsing remitting (0 vs 6) and more frequent dysautonomia (7 vs 1). The demyelinating features were more marked in primary CIDP group and had better outcome compared with secondary CIDP. Conclusions: Of the total patients with CIDP, 38.5% of patients had secondary CIDP which was associated with progressive course, less demyelinating features, and worse prognosis.
Keywords: Chronic inflammatory demyelinating polyradiculoneuropathy, electrodiagnosis, idiopathic, outcome, polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes, secondary, treatment
Chronic inflammatory demyelinating polyradi-culoneuropathy (CIDP) is a syndrome with many causes. Acquired demyelinating polyradiculoneuropathy meeting the diagnostic criteria of CIDP may be associated with HIV infection, systemic lupus erythematosus, chronic active hepatitis, inflammatory bowel disease, Hodgkin's lymphoma, monoclonal gammopathy of undetermined significance (MGUS), and plasma cell dyscrasias such as macroglobulinemia, osteosclerotic myeloma, and Castleman syndrome. ,,,,,,, There may be some clinical and laboratory differences between the patients with idiopathic and secondary CIDP such as those with MGUS are older, have a more protracted course or milder illness at presentation, and poorer therapeutic response. Multiple myeloma and vasculitis on the other hand may have a more severe illness with poorer outcome. Diabetic patients are reported to be 10 times more prone to develop CIDP and have good response to IVIg (intravenous immunoglobulin) or corticosteroid therapy. , The concept that CIDP occurs 10 times more frequently in diabetics has recently been challenged.  There is paucity of report evaluating the clinical spectrum and outcome of various subgroups of CIDP.  During the past 9 years, we have managed 65 patients with CIDP and in this communication we report their clinical spectrum and compare the clinical, laboratory, and prognostic variables in primary and secondary CIDP.
Consecutive patients with CIDP diagnosed as per European Federation of Neurological Societies/Peripheral Nerve Society criteria  during 2001-2009 were included. Detailed clinical history including duration of illness, sensory motor symptoms, gait instability, autonomic dysfunction, as well as symptoms of associated medical illness such as bone pain, edema, jaundice, anemia, gastrointestinal bleeding, diabetes mellitus, thyroid disorder, arthritis, underlying malignancy, and skin pigmentation were noted. CIDP patients with relevant underlying disease were classified as secondary CIDP and those without were classified as primary CIDP. Patients having paraproteinemia were further diagnosed as POEMS (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes), multiple myeloma, and MGUS. ,
The disability of the patients was graded on a 0-10 scale: 0 = normal, 1 = minor sensory signs and areflexia; 2 = minor disability with ambulation more than 200 meters, mild weakness, and sensory impairment; 3 = moderate disability, ambulation more than 50 m, weakness MRC (Medical Research council) grade 4, and sensory impairment; 4 = severe disability, ambulation more than 10 m with support, weakness MRC grade 4, and sensory impairment; 5 = needs support to walk 5 m, marked motor and sensory signs; 6 = inability to walk 5 m but can stand unsupported, able to transfer to wheel chair, and feed independently; 7 = bedridden, severe quadriparesis, MRC grade 3; 8 = respirator and/or severe quadriparesis, MRC grade 2; 9 = respirator and quadriplegia; 10 = dead. 
Presence of cranial nerve palsy and Horner's syndrome were noted. Muscle power was graded on a 0-V MRC scale. Pinprick, joint position, and vibration sensations were noted. Dysautonomia, palpitation, sweating abnormality, diarrhea, and impotence were also noted. The autonomic function was evaluated by respiratory variation of RR interval, postural hypotension, blood pressure response to isometric contraction, and cold immersion test. Abnormality in at least two autonomic function tests was needed to define dysautonomia.  General physical examination was carried out according to a preset protocol.
Hemoglobin, blood counts, ESR (erythrocyte sedimentation rate), bone marrow, serum B12, blood urea, serum creatinine, liver function tests, HIV (Human immune deficiency virus) serology, serum protein, albumin, globulin, antinuclear antibody, anti ds DNA, and CSF (cerebrospinal fluid) examination for protein and cell were carried out in all the patients. Fasting and postprandial (PP) blood sugar and HbA1c were estimated. Fasting blood sugar ≥126 mg/dl (7.0 mmol/l) and PP blood sugar ≥200 mg/dl (11.1 mmol/l) were regarded as diabetes as per ADA (American Diabetic Association) criteria 1997.  T3, T4, TSH (thyroid stimulating hormone), and serum electrophoresis for myeloma protein and immunoglobulin levels were estimated. 24-hour urinary protein was also measured. Chest radiograph, electrocardiogram, and abdominal ultrasonography were carried out. Skeletal survey was also done in the patients with paraproteinemia.
Nerve conduction studies were carried out on Neuromatic 2000 Dantec, Denmark. Motor nerve conduction study of median, ulnar, and peroneal nerves using standard technique was done. The decay of amplitude on proximal stimulation compared with distal if exceeded 50% was regarded as conduction block.  Minimal F latency was elicited in the above mentioned nerves. Orthodromic sensory nerve conduction study of median and ulnar and antidromic sural conductions were carried out. The results of nerve conduction studies were compared with our laboratory's normal value. 
All the patients with moderate to severe disability received prednisolone 0.75 to 1 mg/kg. The patients who did not improve by 3 months were treated by azathioprine 2 mg/kg/day while prednisolone dose was reduced by 50%. The patients with secondary CIDP received treatment of the underlying disease; for POEMS syndrome, monthly cyclophosphamide (15 mg/ kg)  intravenously was given for six pulses. Patients with diabetes were managed by insulin. The patients were followed up at 3 months, 6 months, and later if possible. The patients were considered improved if there was at least two grades improvement in clinical grading scale.
For statistical analysis, the patients were categorized into primary or secondary CIDP based on clinical and laboratory findings. For comparison of different subsets of secondary CIDP, they were categorized into group 1 (diabetes and MGUS) and group 2 (POEMS, multiple myeloma, and malignancy). The clinical severity, outcome, and neurophysiological parameters were compared between primary and secondary groups using Chi square, Fisher's exact, independent 't' test, and one way ANOVA employing SSPS 15 version software.
There were 65 patients with CIDP whose median age was 45 (range, 17-72) years and 20 were females. The median duration of illness was 6 (range, 2-108) months. All the patients had mild to severe weakness and their clinical disability scale (CGS) score range between 2 and 7. The weakness started distally in 38, proximally in 20, and both proximal and distal in seven patients. The sensory symptoms in the form of tingling, numbness, and paresthesia were present in 63 patients. Cranial nerves were involved in nine patients and included third nerve in one, seventh in one, ninth and tenth in eight, and 12 th in two patients. Autonomic dysfunction included postural hypotension in seven, sweating abnormality in two, and skin changes in three patients. Seven of these patients had secondary CIDP and included diabetes mellitus in three, MGUS in two, POEMS in one, and malignancy in one patient. Only one patient with primary CIDP had autonomic dysfunction. Skin pigmentation was present in seven, white nail in three, hepatomegaly in seven, splenomegaly in two, pedal edema in two, and lymphadenopathy in seven patients.
Nine patients were anemic (Hb <12 g/dl) and ESR was high in 21 patients. Sixteen patients were diabetic, four had renal dysfunction (creatinine >1.6 mg/dl), and nine had hepatic dysfunction. In seven patients, TSH was high and in seven patients, electrophoresis revealed M-band. Serum IgG, IgM, and IgA were elevated in nine, three, and two patients, respectively. None of the patients had positive HIV serology and vasculitic profile.
In 25 patients, secondary causes could be found and included diabetes mellitus in 16, POEMS syndrome in two, MGUS in two, lymphoma in one, multiple myeloma in one, and gastrointestinal malignancy in one patient. Hypothyroidism was present as a part of POEMS in four, and was associated with diabetes in one and malignancy in two patients. In the remaining 40 patients, no cause could be established and these patients were included in the primary CIDP group [Figure 1].
Comparison of primary and secondary chronic inflammatory demyelinating polyradiculoneuropathy
On comparing the clinical and laboratory parameters in primary and secondary CIDP, the secondary CIDP patients were older (P=0.03), more frequently anemic (P=0.01), had less frequently relapsing and remitting course (0 patients vs six patients; P=0.07), and more frequently dysautonomia (P=0.004). The comparison between primary and secondary CIDP is presented in [Table 1]. On comparing the neurophysiological findings, the primary CIDP patients had more marked features of demyelination as evidenced by prolonged distal motor latency in peroneal (P=0.0001) and in ulnar nerves (P=0.006). The details of neurophysiological findings in primary and secondary CIDP are presented in [Table 2].
The mean follow up of the patients was 10.6 ± 9.7 months (range, 3 to 54 months). Two patients were lost from follow up. Four patients had follow-up at 3 months, 42 at 6 months, and 17 at 12 months or more. The duration of follow-up was similar in various subgroups. Fifty-eight patients received only prednisolone, two received prednisolone plus azathioprine, and four received prednisolone plus cyclophosphamide pulse monthly for 6 months followed by prednisolone. Four patients with POEMS received local radiation. Within 6 months of treatment, three patients died; two in secondary CIDP and one in primary CIDP group. The underlying diseases in these patients were lymphoma in one, gastrointestinal malignancy in one, and one patient of primary CIDP died as a result of severe sepsis. Of the remaining 62 patients, improvement at 6 months was noted in 48 patients and their mean CGS score improved from 5.60 to 1.98. The improvement was more marked in primary CIDP group.
On comparing the outcome between idiopathic, group 1 secondary (diabetic + MGUS), and group 2 secondary (POEMS and malignancy group) patients, idiopathic CIDP patients improved better at 6 months and at last follow-up compared with group 1 and group 2. The details of the subgroup analysis are presented in [Table 3].
In our study, 61.5% patients had idiopathic CIDP and 38.5% had secondary CIDP which included diabetes mellitus in 24.6%, MGUS in 3.1%, POEMS in 6.2%, and malignancy in 4.6% patients. The higher frequency of diabetics in our study might reflect the higher prevalence of diabetes in India (reported 3.3 to 12%) , compared to the West (1-2%).  The patients with secondary CIDP had progressive course, older in age at disease onset, and had less pronounced demyelinating changes compared with idiopathic CIDP. In a study from France, 68.8% CIDP patients were idiopathic and 31.2% were due to MGUS. The secondary CIDP group had progressive course and was resistant to corticosteroid therapy. The demyelinating features on nerve conduction study such as slowing of nerve conduction velocity, conduction block, and prolongation of distal motor latency were more marked in MGUS group compared to idiopathic.  In another study, idiopathic and paraproteinemia-induced CIDP had similar clinical and neurophysiological changes.  The difference in the results may be due to underlying causes of secondary CIDP. Majority of our patients with secondary CIDP had diabetes mellitus and remaining had MGUS, POEMS, or malignancy. Diabetes usually results in axonal neuropathy. The patients with secondary CIDP are likely to have a progressive rather than relapsing remitting course, whereas the patients in idiopathic group may have either progressive or relapsing remitting course. In a study, idiopathic CIDP revealed more pronounced demyelinating features on electrodiagnosis and better response to corticosteroids compared with diabetic CIDP.  In another study on 31 patients with CIDP, 16 had diabetic, seven had MGUS, and eight had idiopathic CIDP.  The patients with MGUS had poorer response to treatment, whereas diabetic CIDP had better response to IVIg. None of our patients with diabetic CIDP received IVIg because of high cost and they were treated with corticosteroid instead. Their blood sugar levels were controlled with insulin and they responded similar to idiopathic CIDP group. Seven patients with POEMS and multiple myeloma received cyclophosphamide pulse. Improvement of patients with POEMS was similar to idiopathic CIDP. The course of POEMS syndrome is chronic and patients typically survive for more than a decade (median, 13.7 years) in contrast to multiple myeloma.  We have used azathioprine in idiopathic CIDP as steroid sparing or in steroid nonresponders. None of the idiopathic CIDP received cyclophosphamide in our study. In a study on 15 patients with idiopathic CIDP, cyclophosphamide resulted improvement in eleven and worsening in one patient.  Currently, three modalities of treatment for CIDP are available; corticosteroid, IVIg, and plasma exchange. Each of these treatments results in similar improvement ranging from 40% to 95%. The combination therapy results in greater and more frequent improvement. ,, 13.8% (nine patients) of our patients had cranial nerve palsy (four idiopathic, five diabetic). In a Canadian study, the cost effectiveness of IVIg and corticosteroid in the treatment of CIDP concluded that IVIg could not be perceived as a cost-effective therapy of CIDP.  In the developing countries like India, IVIg cannot be afforded by majority of the patients where medical insurance is not widely available. In view of our results, corticosteroid can be safely used with due precautions even in diabetic CIDP. Autonomic dysfunction is a feature of AIDP but presence of it in CIDP suggests secondary cause, especially diabetes.
The improvement at 6 months and last follow-up was better in idiopathic CIDP as compared with diabetic, POEMS, and malignancy. Diabetic demyelinating neuropathy although thought to be immunological basis, but underlying metabolic changes may result in superadded axonal changes, the outcome of which may not be as good as idiopathic CIDP. The more pronounced demyelinating changes in idiopathic CIDP suggest a better outcome.
The present study highlights that about 38.5% patients with CIDP have underlying secondary cause which determines the severity and outcome. Corticosteroid may be a more cost-effective alternative to IVIg in diabetic CIDP.
We thank Mr. K S Bisht, S P Singh for technical help and Mr. Rakesh Kumar Nigam for secretarial help.
[Table 1], [Table 2], [Table 3]