Clinical, electrophysiological and laboratory parameters, and outcome in patients with biopsy proven systemic and nonsystemic vasculitic neuropathy
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.250709
Source of Support: None, Conflict of Interest: None
Aim: To describe the clinical and laboratory findings of patients with biopsy proven vasculitic neuropathy.
Keywords: Electrophysiological evaluation, nerve biopsy, neuropathy, systemic vasculitis, vasculitis
Peripheral neuropathy is a very common disease. In nearly 60% of cases, this manifestation turns out to be of unknown aetiology. Peripheral neuropathy occurs in 50 to 70% of systemic vasculitic syndromes., Vasculitic neuropathy can occur as a primary non-systemic vasculitic neuropathy (NSVN) or secondary to systemic vasculitides (SVS).,,,,, The vasculitic process involves vessels of the size of 50-300μm, which is the calibre of vasa nervosa. Several reports have also described patients with vasculitis restricted to the peripheral nervous system, the so-called non-systemic vasculitic neuropathy., This group needs to be recognized early as it forms one of the types of reversible neuropathies and the quality of life can be improved greatly following its early treatment., In this study, we describe the clinical, laboratory, electrophysiological and histopathological findings of patients in various age groups, evaluated at a single centre with biopsy proven vasculitic neuropathy.
The study was conducted between January 2008- December 2009 with the file review of biopsy proven patients of vasculitic neuropathy. All the patients were identified from the neuropathology archives of the records of patients who had undergone a nerve biopsy. Subjects who had evidence of definite or probable vasculitis were included in the study and a retrospective chart analysis was carried out. The study was approved by the institutional ethics committee.
The details of the clinical symptomatology, neurological deficits, and electrophysiological abnormalities were obtained from case records. The progression of symptoms was classified as acute, of less than 4 weeks duration; sub-acute, of between 4-8 weeks duration; and, chronic, of greater than 8 weeks duration. Results of the laboratory investigations such as the hemogram, erythrocyte sedimentation rate (ESR), biochemistry, serology, and the auto-immune profile, which included the rheumatoid arthritis (RA) factor, antinuclear antibody (ANA), and anti- neutrophil cytoplasmic antibody (ANCA) were recorded, wherever they were available.
Electrophysiological studies were carried out using standard protocols, and at least one motor and one sensory nerve each in the upper and lower limbs were examined. Values beyond 2 standard deviations (SD) of laboratory control data were considered abnormal. Based on the available data, categorization of sensory/sensorimotor, axonal/demyelinating, mononeuritis multiplex/polyneuropathy, and symmetric/asymmetric neuropathies were made.
The term 'mononeuropathy' was referred to as an abnormality exhibited in a single nerve; the term 'mononeuritis multiplex' was designated when more than one nerve but less than 50% of the tested nerves showed abnormalities; and, the term 'polyneuropathy' indicated that more than 50% of the tested nerves showed an abnormality.
Disparity in the same limb was stated to have been present when one nerve exhibited a normal response and another exhibited >50% reduced response compared with the lower limit of normal. A disproportionate upper limb involvement indicated that the upper extremity responses were more abnormal than the lower extremity ones. A side-to-side asymmetry indicated that in those patients in whom bilateral conductions were measured, a more than two fold amplitude difference existed between the two sides. The presence of axonal/demyelinating neuropathy was indicated based upon on the standard diagnostic criteria.,
Biopsy of the sensory nerve was conducted after establishing the electrophysiological abnormality. The nerve biopsies were fixed in 2.5% glutaraldehyde and one portion was processed for paraffin embedding. A second portion was prefixed in Fleming's solution and processed by the Kulchitsky Pal stain for the assessment of myelin. Longitudinal and transverse sections of 3 microns thick were serially cut and stained with the hematoxylin-eosin stain, Masson's trichrome stain for collagen, and Kulchitsky Pal stain for myelin. In addition, periodic acid-Schiff, Congo red, and Perl's Prussian blue staining procedures were carried out for detecting paraprotein/immunoglobulin, amyloid, and hemosiderin deposits, respectively. Immunohistochemistry by the indirect immune peroxidase method was carried out in selected cases using antibodies to leucocyte common antigen (monoclonal 1:100, Bio Genex, Fremont, CA, USA) to detect inflammatory infiltrates.
All nerve biopsy specimens were systematically reviewed by a single neuropathologist for the detecting the presence of sub-perineurial edema, myelinated fibre loss, acute myelin/axonal breakdown, demyelination, axonal regeneration, inflammatory cell infiltrate and vascular alterations. Biopsies were diagnosed as having a definite, probable, or possible vasculitis, in accordance with the Collins' criteria. The criteria included the following:
Definite vasculitis: Presence of transmural inflammatory infiltration with or without fibrinoid necrosis; and, Probable vasculitis: Presence of at least one vessel rimmed or infiltrated by inflammatory cells, in the absence of transmural infiltration; and, the presence of other supportive pathologic features, namely either vascular alterations (vascular thickening and sclerosis, narrowing or obliteration of lumen, thrombosis with or without recanalization, epineurial capillary proliferation or neovascularization and periadventitial hemosiderin deposits) or asymmetric nerve fibre loss and active Wallerian-like degeneration. In this study, inflammation was considered significant if ten or more epineurial inflammatory cells or five or more endoneurial inflammatory cells were present. A minimum of two vascular changes in the presence of inflammation and sectorial myelin loss was considered significant.
Possible vasculitis: The same criteria as for probable vasculitis existed, except for a milder degree of inflammation in less than five cells in the endoneurial, and 5-9 cells in the epineural compartment.
Additional tissue biopsies were reviewed wherever available. In the present study, only cases having either a definite or a probable vasculitis were included.
Data on treatment and follow-up evaluation were gathered from the case records. The nature of the immunosuppressants used, the response to treatment, and the duration of follow up were noted. The outcome was assessed in those patients having a follow-up duration of at least 6 months and was categorized as being either 'good' or 'poor'. A 'good' outcome indicated an asymptomatic patient/a patients with no disability/a patient independent for all activities of daily living; a 'poor' outcome indicated a patient who needed help in day-to-day activities/a patient who needed assistance for walking/a patient who was unable to walk/a patient who died.
Depending on the clinical features, laboratory parameters and histopathology, the patients were grouped into the categories of systemic vasculitis (SVN) and non-systemic vasculitic neuropathy (NSVN). In the presence of constitutional symptoms alone, without any other evidence of systemic involvement (apart from involvement of nerve/muscle), the patients were classified as having non-systemic vasculitis. In the presence of multiorgan involvement but without features satisfying the criteria of any defined systemic vasculitic syndrome, the cases were labelled as undifferentiated/unclassified vasculitis.
Statistical analysis was done with the Statistical Package for the Social Sciences (SPSS) software. Frequencies were calculated for categorical variables and means for continuous variables. The chi square test/Fisher exact test were used for comparing the categorical variables between the groups, and independent sample t test was used for the continuous variables.
During the study period, 76 patients had undergone a nerve biopsy with evidence of definite or probable vasculitis. Nine patients were excluded in view of their incomplete medical records. The data of the remaining 67 patients was analysed. Histopathology revealed the presence of definite vasculitis in 37 patients and probable vasculitis in 30 patients.
Demographic details: The age at presentation ranged from 18 to 69 years (the mean age being 42.67 ± 13.56 years) with the majority (51%) of subjects being in the fourth and fifth decades [Figure 1].
There was no gender predilection, with both male and female patients being almost equally affected (male: female = 33:34) [Figure 2].
The onset of neuropathic symptoms was in the lower limbs in the majority (86.6%) of patients. The initial neuropathic symptom was the presence of paresthesias in 68.7% of patients; and, the presence of both paresthesias and weakness in 28.4% patients [Table 1]. The most common clinical pattern was polyneuropathy, seen in 70.1% patients; and, mononeuritis multiplex, seen in 29.9% of patients. A pure sensory neuropathy was seen in 14 patients. The clinical course was chronic in 54 (80.59%), sub-acute in 8 (11.94%) and acute in 5 (7.46%) patients.
An elevated erythrocytic sedimentation rate (ESR) was observed in 80.59% patients, and half of them had an ESR value of more than 50mm/hr. The rheumatoid arthritis (RA) factor was positive in 19 of 59 patients. 7 out of 59 patients tested positive for antinuclear antibodies (ANA). Ro 52 was positive in 4 out of 29, ds DNA (double stranded deoxyribose nucleic acid) in 2 out of 29, anti- Sm (anti-Smith antibodies) in 2 out of 29, anti-RNP (ribonuclear protein) in 1 out of 29, p-ANCA (perinuclear anti-neutrophil cytoplasmic antibodies) in 1 out of 25 patients. Three patients were found to be human immunodeficiency virus (HIV) reactive, one of whom was also HBsAg (surface antigen of the hepatitis B virus) reactive [Table 2].
Of the 67 patients, bilateral nerve conduction studies were available in 38 patients and remaining 29 had unilateral motor and sensory studies. Axonopathy was seen in 95.52%, and a predominant demyelinating neuropathy only in 4.48% of patients. Sensorimotor neuropathy was observed in 83.58%, and a pure sensory neuropathy was seen in 16.42% patients. More than two-thirds (67.16%) of the patients had polyneuropathy. Among the patients with bilateral nerve conduction studies, side-to-side asymmetry was also observed in 71.05% patients [Table 3].
Of the 67 patients, sural nerve biopsy was done in 61, superficial peroneal biopsy in 3, and dorsal cutaneous nerve of the ulnar nerve biopsy in 3 patients. The nerve biopsy revealed definite vasculitis [Figure 3] and [Figure 4] in 37, and probable vasculitis [Figure 5] in 30 patients. Rheumatoid arthritis with vasculitis was seen in 7 cases [Figure 6]. The muscle biopsy revealed vasculitis in 9 out of 13 patients, and the skin biopsy revealed vasculitis in 5 out of 7 patients. Lip biopsy confirmed sialadenitis in 3 patients with Sjogren's syndrome [Figure 7], HIV vasculitis in 3 cases [Figure 8], SLE vasculitis in 2 cases [Figure 9], Churg Strauss in 1 case [Figure 10], and systemic sclerosis in 1 case [Figure 11].
Based on the clinical, laboratory and histopathological features, out of 67 patients, 46 had isolated vasculitis of the peripheral nervous system (NSVN), and 21 had systemic vasculitides. Sub-categorization of the systemic vasculitic syndromes is shown in the [Table 4].
The patients received immune-suppressants based on the associated comorbidities and the financial feasibility.
Majority of the patients (60/67) received corticosteroids, either oral prednisolone or parenteral (intravenous [iv] methylprednisolone) followed by oral prednisolone maintenance therapy. Twenty-three patients received intravenous pulse doses of cyclophosphamide. Other medications used included azathioprine (n = 6); methotrexate (n = 4); mycophenolate mofetil (n = 2); hydroxychloroquine (n = 2) and highly active anti-retroviral therapy (HAART) [n = 3].
Follow-up and outcome
Of the total of 67 patients, 49 had minimum of 6 months follow up. 36 of these patients had a good outcome (asymptomatic/independent for all activities of daily living) and 13 (1 patient expired) had a poor outcome (dependent for activities of daily living or death). The mean duration of follow up was 10.90 ± 9.58 months. There was a significant difference in the outcome between systemic vasculitis (43.8% good, 56.2% poor) and nonsystemic vasculitic neuropathy (87.88% good, 12.12% poor).
On comparative analysis between systemic vasculitis and nonsystemic vasculitic neuropathy, the clinical pattern of neuropathy, electrophysiology, and histopathological findings did not differ much. However, the systemic vasculitis group had a significantly higher systemic symptoms, an elevated ESR, anemia, a positive autoantibody profile and a poor outcome [Table 5].
Peripheral neuropathy is a common feature in both systemic and nonsystemic vasculitis as the vasculitic process involves vessels of the size of 50-300μm, which is the caliber of vasa nervosa. Vasculitic neuropathy can occur as an isolated process or as a manifestation of systemic disorder with multi-organ involvement. It accounts for 5.3% to13.8% of biopsy proven peripheral neuropathies.
In this study, nonsystemic vasculitic neuropathy (46/67, 68.66%) was more frequent than systemic vasculitis (21/67, 31.34%). Studies from the West have reported more number of patients with systemic vasculitis than non-systemic vasculitic neuropathy. However, our findings are comparable to the series reported by Murthy et al., which included 56.25% patients of nonsystemic vasculitic neuropathy and 43.75% of systemic vasculitis. The most likely explanation for this disparity is a referral bias. As our centre is a tertiary care centre catering exclusively to neurological and psychiatry disorders, the proportion of cases of systemic vasculitides with multiorgan involvement is expected to be small, as most of these patients would be sent to other multi-speciality hospitals. The mean age at presentation of the cohort was 42.67 ± 13.56 years (range: 18-69 years). This study population was younger compared to the published studies from the Western literature. Collins et al., in their study reported a mean age of 62.1 ± 14.3 years, which is similar to that of Hawke et al. However, our findings (42.67yrs) are comparable to a study from South India in which 42.37 years was the mean age at presentation.
In the present study, twenty-six of sixty-seven patients had systemic (including constitutional) symptoms. The most common systemic symptoms included arthralgia (26.9%), fever (25.4%) and weight loss (13.4%). The presence of systemic/constitutional symptoms is well known in systemic vasculitic neuropathy, but they can occur in nonsystemic vasculitic neuropathy as well. The previous studies have shown that weight loss and fever are the most common systemic manifestations in vasculitic neuropathy., As significant proportion of patients in the present study had vasculitis secondary to connective tissue disorders, arthralgia/arthritis was the most common systemic symptom.
In vasculitic neuropathies, the pattern of neuropathic involvement depends on the extent and temporal progression of vasculitis induced ischemic changes. The duration of neuropathic symptoms at the initial presentation was 9.27 ± 13.73 months, which reflects the chronic course of vasculitic neuropathy and is comparable to the existing literature. Sensory symptoms in the form of painful paresthesias was the most frequent (68.7%) initial neuropathic symptom at presentation; whereas, predominantly motor symptoms at the onset were extremely rare (3%), the presence of which in fact, some authors consider as exclusion criteria for vasculitic neuropathy. Though sensorimotor neuropathy was the most common pattern, almost one-fifth (20.9%) of patients in this study clinically had pure sensory neuropathy. The occurrence of pure sensory neuropathy in vasculitis is well described.,
Elevated ESR greater than 50 mm/hr favours the presence of an inflammatory pathology. Though available in limited number of patients, the RA factor, as well as the ANA and ANCA profile were useful in classifying the vasculitic syndromes. In this study, elevation in ESR was seen in more than 80% of cases, and in 40% of patients, it exceeded 50 mm/hr. This observation concurs with the results reported in the existing literature., The RA factor was positive in 28.8%% and ANA in 12% of patients. The auto-antibody profile was positive in 17% patients. All these laboratory findings differed significantly between the systemic and nonsystemic vasculitic neuropathy subgroups, with the autoantibody profile and ANA being positive only in the systemic vasculitis cohort. The RA factor was, however, positive in 7.5% of the NSVN group. Collins et al., also reported similar observations. Dyck et al., reported a positive RA factor and ANA in a small number of patients with NSVN. This has been attributed to the low specificity of these tests, being positive in several non- rheumatological conditions. It has also been suggested that that it may reflect a subclinical/latent systemic vasculitic syndrome without full blown manifestations, or that isolated peripheral nervous system vasculitis/NSVN may in fact be forme fruste of systemic vasculitis.
Electrophysiologic testing also revealed diffuse polyneuropathy as the most common (67.16%) pattern, followed by mononeuritis multiplex in one-third (32.8%) of patients. Differences in conductions between the ipsilateral median and ulnar nerves were observed in 28.3% of patients. In those cases where nerve conductions were done on both the sides, more than two-thirds (71.05%) of patients demonstrated asymmetry. Axonopathy was evident in 95.5% of patients. Pure sensory nerve affection on electrophysiology was seen in 16.41% of patients. These observations are in accordance with other studies. Zivkovic et al., reported that polyneuropathy was the most common pattern followed by mononeuritis multiplex. Side- to-side asymmetry and disparity in the same limb were seen in 67% and 35% of the patients, respectively. Murthy et al., reported asymmetrical motor sensory neuropathy with mononeuritis multiplex as the most common pattern of neuropathy in both systemic vasculitis and nonsystemic vasculitic neuropathy. Though the classical pattern of neuropathy in vasculitis is mononeuritis multiplex, the relative frequencies of polyneuropathy versus mononeuritis multiplex is found to be highly variable. The likely explanation for this discrepancy could be the late presentation of the patients for the initial evaluation, due to which confluent mononeuropathies would mimic diffuse polyneuropathy. However, there is no uniformity in the terminologies used in different studies, as the standard defining criteria do not exist, making comparisons difficult. Although axonopathy is the characteristic feature of vasculitic neuropathy, focal conduction blocks and demyelinating neuropathy were also described. In some patients, though the clinical examination may suggest a pure sensory neuropathy, electrophysiologic testing revealed subtle abnormalities in motor conductions. Three of our patients had similar findings. There were no significant differences in the electrophysiological findings between patients with systemic and nonsystemic vasculitic neuropathies, a result consistent with the previously reported studies.,
In the present study, sural nerve was most frequently biopsied nerve in 61 (91.04%) patients. Superficial peroneal and dorsal cutaneous branch of the ulnar nerve were biopsied in 3 patients, respectively. The selection of the nerve to be biopsied was based on the clinical and electrophysiological observations.
Of these patients, 37 cases demonstrated features of definite vasculitis on the biopsy, while 30 had probable vasculitis. As the entry point for our study was histopathologically proven vasculitis, the sensitivity of the nerve biopsy in suspected vasculitic neuropathy could not be commented upon.
The muscle biopsy revealed vasculitis in 69% of patients, with peroneus brevis having the highest yield. However, the sensitivity of muscle biopsy could not be ascertained in this study, as the number of cases in which muscle biopsy was performed was limited (n = 13). However, the additional yield of muscle biopsy has been described previously.,, Lip biopsy revealed sialadenitis in three patients of Sjogren's syndrome. The utility of the lip biopsy in diagnosing the condition is well known.
Three patients in the present study had human immunodeficiency virus (HIV) associated vasculitic neuropathy. The nature and patterns of peripheral neuropathy have been described previously., Connective tissue disorders accounted for the significant proportion of patients in the systemic vasculitic neuropathy subgroup. Vasculitic neuropathy is known to occur in various rheumatological diseases.,,,
The response of vasculitic neuropathy to steroids and immunosuppressants is variable. The combination of steroids and cyclophosphamide is recommended based on the available literature. In view of the chronic and relapsing nature of the illness, a drop-out from the treatment arm is expected at sequential follow-up visits, as observed in the present study.
Most (90%) of the patients received parenteral or oral steroids. One-third of them received cyclophosphamide. Three patients with HIV related vasculitis received highly active antiretroviral therapy (HAART) alone. Azathioprine and methotrexate were used in a small fraction (9% and 6%, respectively) of patients. In one of the largest series of vasculitic neuropathy, 54% of patients received cyclophosphamide. Collins et al., advocated a combination regimen (steroid and cyclophosphamide) even in nonsystemic vasculitic neuropathy.
Overall, the outcome was good in 73.46% of patients There was a significant difference between the systemic (43.8%) and nonsystemic vasculitic neuropathy (87.88%) subgroups. Mathew et al., reported (in the subset of patients with positive nerve biopsies), an overall good outcome in 76% of those patients treated with cyclophosphamide and steroids. In their study, all the patients of nonsystemic vasculitic neuropathy had a good outcome irrespective of the treatment regimen. However, Collins et al., reported that in nonsystemic vasculitic neuropathy, a combination of steroid and cyclphosphamide was more effective than steroids alone, which was consistent with the result of the observational studies in systemic vasculitis. Murthy et al., reported a good outcome in 8 out of 9 patients of nonsystemic vasculitic neuropathy, and in 5 out of 7 patients of systemic vasculitis. Corticosteroids were used in fifteen patients, and only one patient received cyclophosphamide. Hawke et al., also reported a poor long-term prognosis in systemic vasculitic syndromes.
These apparent discrepancies in outcome with various regimens may be related to the difference in treatment protocols and a bias towards using cyclophosphamide in those patients with severe disease as against steroids alone in those with milder disease, which would influence the final outcome. Randomized studies comparing the steroid alone regimen versus a combination therapy may better address this issue.
The strength of the present study has been in gathering histopathological evidence of definite or probable vasculitis on nerve biopsy as the entry point for the patients in the study. Histopathology is the gold standard diagnostic test for vasculitic neuropathy, and the exclusion of possible vasculitis where an alternative diagnosis is present, may be possible. This is one of the largest series from a single centre reported till date. Its major limitations included the fact that the immunological profile was available in a limited number of patients; and, bilateral nerve conduction studies in the entire cohort could have strengthened the observations. In view of the retrospective nature of the study, the treatment and follow-up protocols were not uniform.
Vasculitis is a potentially treatable etiology of peripheral neuropathy. The prognosis is good in isolated vasculitis of the peripheral nervous system when compared to systemic vasculitides. The timely diagnosis of these conditions is important for planning treatment protocols, educating patients regarding their medication compliance and adverse drug reactions and for adequately following-up these patients. Prospective studies may better address the few limitations of our study.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8], [Figure 9], [Figure 10], [Figure 11]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]