A 73-year-old woman developed malaise and a wasting illness that led to a weight loss of 80 pounds over the 5 years before presentation. During this time, her body mass index declined from 30.3 to 17.0 kg/m2 (1). At approximately the same time her weight loss began, the patient began to experience chronic pelvic and bladder pain. These symptoms, particularly uncomfortable when she was sitting, were partly relieved by micturition. The pain was associated with intermittent dysuria, severe urinary urgency, and frequency. She urinated approximately 9 times a day, and also awoke to urinate approximately 9 times at night.
She denied incontinence and did not have recurrent urinary tract infections. However, her symptoms were associated with dyspareunia. She was not sexually active because of pain. Gynecologic and urologic evaluations revealed that the external genitalia, perineum, anus, and rectum were normal, as was the strength of the pubococcygeal and external anal sphincter muscles. Her gynecologist had detected mild urethral tenderness on palpation, but there was no urethral or bladder prolapse. The cul-de-sac between the posterior vagina and anterior rectum was normal, without nodularity or an enterocele.
Catheterized urine specimens showed reactive uroepithelial cells, many neutrophils, and some red blood cells, suggesting abundant acute inflammation. A computed tomography (CT) scan of the pelvis showed generalized thickening of the bladder wall (Figure 1). A cystoscopic examination revealed no uroepithelial masses, but demonstrated a trabeculated bladder surface. Biopsies obtained at cystoscopy revealed nonspecific bladder inflammation. No neoplasm was identified.
Following these urologic evaluations, the patient was diagnosed with interstitial cystitis/painful bladder syndrome (IC/PBS) (2). Over the next 5 years, she was treated with a variety of medications, including neurontin, pentosan polysulfate sodium, oxybutynin, and dimethyl sulfoxide. She also underwent cystoscopy with hydrodistention. None of these interventions provided more than mild, temporary relief of her bladder symptoms.
In this same 5-year period, the patient had a variety of persistent abdominal symptoms. She reported loose bowel movements but denied diarrhea, constipation, hematochezia, fevers, arthritis, or eye problems. She noted consistent anorexia and early satiety, and reported a band-like abdominal pain that started just above her umbilicus. She had grown increasingly weak since becoming ill, and usually required assistance for ambulation.
Over the course of her abdominal symptoms, she had undergone upper and lower endoscopy, as well as sonograms and CT scans of her abdomen. Biopsies of the small intestine had been negative for Tropheryma whipplei. Assays for IgA anti-endomysial antibodies, IgA antibodies to transglutaminase, and IgG and IgA directed against gliadin were all negative. A variety of diagnoses for her gastrointestinal symptoms had been entertained, including the postcholecystectomy syndrome, lactose intolerance, and irritable bowel syndrome. She was treated with alosetron hydrochloride. Her weight loss, which occurred at a steady rate during the 5 years leading up to her presentation, had resulted in a decline in her weight from more than 180 pounds to approximately 103 pounds.
Approximately 2 years after the onset of her chronic pelvic pain, abdominal symptoms, and weight loss, the patient had noted progressive shortness of breath. Over a period of several months, the patient's dyspnea progressed such that she became winded after walking only 1 block. In describing this symptom, she distinguished clearly between shortness of breath and fatigue. Her shortness of breath was accompanied by a dry cough. She denied fevers, wheezing, and hemoptysis. A chest radiograph showed diffusely increased interstitial markings at the lung bases (Figure 2A). A CT scan of the chest was consistent with pulmonary fibrosis (Figure 2B), but these findings were not evaluated further. She was given the tentative diagnosis of idiopathic pulmonary fibrosis, but was not started on any type of therapy. Over the next 3 years, serial chest radiographs indicated that her pulmonary process remained stable. However, her overall condition continued to decline through additional weight loss and progressive weakness.
In 2005, during the 1 year preceding her presentation to a rheumatologist, the patient developed the subacute onset of a burning sensation in her toes. The specific descriptions of this pain ranged from “prickling” to “feeling like my feet were on fire.” These distal dysesthesias, most prominent at the end of the day, were exacerbated by prolonged periods with her legs in a recumbent position. They worsened greatly at night, and prevented the patient from sleeping more than 3 consecutive hours. A neurologic examination revealed normal power and tone in the proximal and distal extremities. The deep tendon reflexes were intact. Proprioception was normal, but pinprick sensation was decreased in a stocking gradient.
Several glucose tolerance tests in the period leading up to her presentation were diagnostic of glucose intolerance but not frank diabetes, revealing 2-hour serum glucose values >140 mg/dl but <200 mg/dl. A serum protein electrophoresis revealed a polyclonal gammopathy, but immune electrophoresis identified no monoclonal proteins. A urine protein electrophoresis was normal. Vitamin B12 level was 713 pg/ml (normal range 200–900). A lumbar puncture revealed no white blood cells, and the cerebrospinal fluid protein was normal (32 mg/dl). There were no oligoclonal bands, and the cerebrospinal fluid IgG index was 0.2 (normal range 0.2–0.8).
The erythrocyte sedimentation rate was 76 mm/hour (normal <20). The C-reactive protein level was 2.8 mg/dl (normal <0.5). In addition, a serologic evaluation ordered by the neurologist revealed antinuclear antibodies at a titer of 1:80 (speckled pattern) and a strongly positive rheumatoid factor (RF; 574 IU, normal <20). Electrodiagnostic tests of the neuromuscular system revealed normal electromyography and nerve conduction velocities of both the upper and lower extremities. To evaluate the possibility of a small fiber neuropathy, the patient underwent skin biopsies of the thigh and distal leg. The skin biopsy revealed severely decreased nerve fiber density in the epidermis (Figure 3A). Many nerve fibers did not extend beyond the dermal–epidermal junction, and the distribution of nerve fibers was patchy. In healthy controls who do not have small fiber neuropathy, nerves innervate the epidermis abundantly, typically with a varicose appearance (Figure 3B). Our patient's biopsy samples were consistent with a moderately severe, length-dependent process that affected peripheral nerves, with involvement restricted to the small-caliber sensory fibers.
Past medical history
The patient had a history of osteoporosis, total knee replacement, transvaginal hysterectomy, laparoscopic cholecystectomy, and bilateral cataract removal. She was gravida III, para II, and had experienced 1 spontaneous abortion in the first trimester of pregnancy.
Her medications included raloxifene and calcium carbonate for her osteoporosis, and oxybutynin, neurontin, and pentosan for her bladder symptoms.
Family and social history
The family history was noncontributory. She was married and had 2 grown children. She had a remote history of tobacco use and rarely drank alcohol.
Review of systems
The patient reported occasional frontal headaches, relieved with acetaminophen. She denied scalp tenderness, visual changes, pain in the jaw with chewing, and symptoms of polymyalgia rheumatica. She also denied arthritis, oral ulcers, alopecia, dry eyes, dry mouth, Raynaud's phenomenon, and skin rash.
The patient was a cachectic, weak-appearing woman. She weighed 102.1 pounds and was 165 cm tall (body mass index 17.0 kg/m2). Her temperature was 35.9°C, blood pressure 134/87 mm Hg, respiratory rate 12 breaths/minute, and heart rate 110 beats/minute. The head, eye, ear, nose, and throat examination was remarkable for severe temporal wasting, but there were no oral or nasal ulcers. Her mucous membranes were moist, and her salivary pool adequate. There was no lymphadenopathy or thyromegaly. The cardiovascular examination showed a regular rhythm without murmurs, rubs, or gallops. There were no bruits over the large vessels of the upper or lower extremities. Auscultation of the lungs revealed dry crackles halfway up both lung fields. Her abdomen revealed a transverse scar several centimeters below the umbilicus. There was no organomegaly, skin rash, cutaneous nodules, synovitis, or joint deformities. Neurologic examination revealed intact cranial nerves, normal proximal and distal power in the upper and lower extremities, and preserved patellar and Achilles tendon reflexes. Proprioception was normal, but the pain (pinprick) and temperature modalities were moderately decreased distal to the mid-shins. Vibratory sensation was mildly decreased in the toes.
Results of the routine laboratory examinations are shown in Table 1. A second RF assay was positive (300 IU/ml, normal <36), but an assay for antibodies directed against cyclic citrullinated peptides was negative. An antinuclear antibody assay was positive at a titer of 1:40. However, antibodies to double-stranded DNA were negative, as were those to extractable nuclear antigens (Ro, La, Sm, RNP). The serum C3 level was elevated slightly (155 mg/dl, normal range 79–152), but the C4 level was normal (38 mg/dl, normal range 12–42). Assays for anti-Jo1 and antimitochondrial antibodies were negative, but the anti–smooth muscle antibody was positive at a titer of 1:40. An immunofluorescence assay for antineutrophil cytoplasmic antibodies (ANCAs) was positive at a titer of greater than 1:160 in a perinuclear pattern. An enzyme immunoassay for antibodies to myeloperoxidase was positive at 31.6 units (moderately to strongly positive at <30.0). Antibodies to serine proteinase-3 were negative.
Table 1. Results of laboratory examinations
White blood cell count, per mm3
Platelet count, 1,000/mm3
Mean corpuscular volume, femtoliters
Red blood cell distribution width, %
Prothrombin time, seconds
Partial thromboplastin time, seconds
Urea nitrogen, mg/dl
Alkaline phosphatase, units/liter
Aspartate aminotransferase, units/liter
Alanine aminotransferase, units/liter
Red blood cells
White blood cells
The patient is an elderly woman with a 5-year history of chronic bladder and pelvic pain diagnosed as IC/PBS, an 80-pound weight loss, a small fiber neuropathy, and interstitial lung disease. She has constitutional symptoms, a strongly positive RF, and positive ANCA assays associated with perinuclear immunofluorescence and antigen specificity for myeloperoxidase.
We review the differential diagnosis from the standpoint of the major organ systems involved in the patient's illness.
Genitourinary differential diagnosis
The patient's most persistent symptoms were related to her lower genitourinary tract. The consensus among her urologists was that she had idiopathic IC/PBS (2), which is typified by the classic symptom of bladder pain relieved by voiding. The International Continence Society definition of this challenging clinical syndrome refers to the “complaint of suprapubic pain related to bladder filling, accompanied by other symptoms such as increased daytime and nighttime frequency in the absence of proven urinary infection or other obvious pathology.” The label of IC/PBS should be applied only after the exclusion of other conditions that cause similar symptoms, including bacterial cystitis, bladder cancer, urethral diverticulum, sexually transmitted diseases, bladder stones, ureteral stones, radiation cystitis, chemical cystitis (e.g., cyclophosphamide-induced bladder injury), and tuberculous cystitis. Therefore, IC/PBS is a diagnosis of exclusion.
The treatment of IC/PBS, empiric in nature, is directed at relief of most troublesome symptoms. Many treatments have proven ineffectual when subjected to randomized trials (3–5). Before we evaluated the patient, she had received and failed all of the standard interventions for the treatment of this condition.
Neurologic differential diagnosis
Patients with small fiber sensory neuropathies typically report pain. Pain in the feet and soles is the dominant symptom among patients with distal small fiber loss. In contrast, for patients with large fiber neuropathies, numbness is the predominant symptom. The severity of neuropathic symptoms in small fiber neuropathy can range from mildly annoying to incapacitating. Descriptors such as “tingling,” “prickling,” “burning,” “on fire,” or “electric shocks” are used to describe these pathologic sensations. Despite these compelling histories, patients with small fiber sensory neuropathies often have normal neurologic examinations. Small fiber nerves cannot be assessed by deep tendon reflexes or the standard neurologic examination techniques of proprioception or vibration testing. Furthermore, patients with neuropathies that primarily involve small nerve fibers usually have normal results of electromyogram/nerve conduction velocity testing, which was the case with our patient.
Small fiber neuropathy presents with a length-dependent, usually symmetric, pattern of nerve dysfunction that has a strong predilection for the distal extremities. The lower extremities are affected first. Small fiber nerve injury is increasingly recognized as a cause of the syndrome of painful neuropathy (6, 7). The pain associated with this disorder is usually described by terms relating to numbness, burning, or shooting pain. Damage to the smallest myelinated fibers, the A-alpha fibers, sharply mediates localizable burning pain. In contrast, damage to unmyelinated nerves, the C fibers, causes poorly localizable numbness (8). The differential diagnosis for small fiber neuropathy includes familial or acquired amyloidosis (9), Sjögren's syndrome (10), paraneoplastic neuropathy (11), and diabetes mellitus (12).
Although the small fibers are affected in 15–30% of patients with amyloidosis, a monoclonal gammopathy is detected in 90% of such individuals, and autonomic symptoms are also usually prominent (9). Our patient had neither of these features. Moreover, features of a large fiber neuropathy gradually accumulate in amyloidosis and can be diagnosed with nerve conduction velocity studies. Our patient had no evidence of a large fiber neuropathy.
Sjögren's syndrome can present either with features of a small fiber neuropathy or features of an immune-mediated attack against sensory ganglia (i.e., a ganglionopathy), with loss of tendon reflexes and reduction of sensory nerve action potentials (11). However, the absence of both sicca symptoms and autoantibodies to the Ro and La antigens makes this diagnosis unlikely in our patient. Paraneoplastic small fiber neuropathy, which often occurs simultaneously with a proximal ganglionopathy or encephalomyelitis, is often accompanied by anti-Hu antibodies or other autoantibodies (13). In our patient, neither clinical nor electrodiagnostic features of a paraneoplastic ganglionopathy or encephalomyelitis were apparent.
Finally, although our patient had glucose intolerance, which is associated with small fiber neuropathy in up to 39% of patients (12), the other systemic features of this patient's illness demand a broader explanation.
Gastrointestinal differential diagnosis
The patient's progressive, unexplained weight loss over 5 years had been evaluated extensively. One possibility was that the patient's gastrointestinal symptoms were indirectly related to her IC/PBS. It was also possible that her weight loss was the result of a systemic illness and not primarily the manifestation of a gastrointestinal problem.
Celiac sprue can be associated with significant weight loss syndromes that affect the peripheral (and central) nervous system. Although the classic disease features pertain to symptoms and signs of malabsorption, atypical presentations are common. Celiac sprue can be associated with ataxia and/or peripheral neuropathies (14), but the neuropathy is predominantly an axonal, larger fiber neuropathy that seldom affects small nerves. Inflammatory bowel disease has also been more commonly associated with a large fiber neuropathy than a small fiber neuropathy. Furthermore, hypovitaminosis A and E related to malabsorption can cause a neuropathy in celiac sprue and inflammatory bowel disease (15). However, the absence of melenic stools, hematochezia, and steatorrhea, as well as her extensive gastrointestinal evaluations that included small bowel biopsy, effectively exclude these diagnoses.
Whipple's disease, a systemic disorder for which the etiology eluded detection for decades after its original description, was recognized in the 1990s to be caused by Tropheryma whipplei (16). Multiplication of the causative organism within the lamina propria of the small intestine leads to fat malabsorption and weight loss. Neurologic complications of Whipple's disease include dementia, supranuclear gaze palsy, and movement disorders (17), but peripheral neuropathy is distinctly unusual. Moreover, the patient had had a small bowel biopsy that was negative.
Pulmonary differential diagnosis
Interstitial lung disease causes dyspnea through the impairment of oxygen transport from alveoli to red blood cells. Our patient's lung problem had been termed idiopathic pulmonary fibrosis (IPF), a subset of idiopathic interstitial pneumonia associated with a pattern of bibasilar fibrosis, typically relentless progression, and death within 2 years of the diagnosis (18). The relative stability of our patient's lung problems is not typical of IPF.
Many rheumatic syndromes can be associated with pulmonary fibrosis (19). Our patient's pattern of pulmonary fibrosis that involved the apices as well as the basilar portions of the lung is uncharacteristic of IPF, but quite typical of systemic inflammatory disorders such as systemic sclerosis, inflammatory myopathy, vasculitis, and others (19, 20).
Before the patient's rheumatologic evaluation, the refractory nature of her presumed IC/PBS had led to the performance of a palliative cystectomy with ileal loop urinary diversion. This accounted for the transverse scar on her lower abdomen. Surgical pathology of the bladder revealed necrotizing arteritis involving the medium-sized vessels of the bladder wall, specifically within the muscularis propria and perivascular adipose tissue. Pathology of the left ureter also showed necrotizing arteritis within the periurethral tissue (Figure 4). The finding of a medium-vessel vasculitis in these tissues led to the diagnosis of microscopic polyangiitis (MPA).
MPA, a form of necrotizing vasculitis, can have dramatic and life-threatening presentations in different organ systems. When presenting as a pulmonary-renal syndrome, leukocytoclastic capillaritis can lead to alveolar hemorrhage and a rapidly progressive, crescentic glomerulonephritis (21). The diagnosis of MPA is not difficult in the presence of such classic types of organ involvement. However, MPA can also present with more indolent symptoms in organs that are either typical or atypical of MPA. Our patient's bladder involvement falls into this category.
The most striking features of our patient's presentation were her intensely painful bladder symptoms that led to cystectomy, her troubling small fiber neuropathy, and her dramatic weight loss. In retrospect, the nonspecific abdominal pains and weight loss in our patient were systemic effects of her underlying MPA. Genitourinary manifestations of ANCA-associated vasculitides are rare, but described (22). Approximately 50% of patients diagnosed with interstitial cystitis have received incorrect diagnostic labels (23). It is likely that a deep bladder biopsy, which would have required the use of a resectoscope, would have revealed the vasculitis of medium-sized vessels. However, the standard approach to the evaluation of possible IC/PBS does not include biopsies of such depth, which carry with them some additional potential morbidity. The usual approach to cystoscopic biopsies is to perform deep procedures only if a bladder malignancy is suspected. One of the lessons of our case is that for patients with evidence of a multiorgan system process in which the diagnosis is not clear, a deep bladder biopsy might yield diagnostic tissue if there is not a more obvious organ to biopsy.
The most widely-recognized peripheral nerve manifestation of MPA is mononeuritis multiplex. Mononeuritis, caused by pathology within the vasa nervorum of named nerves, is a large fiber axonal neuropathy (24). Mononeuritis multiplex affects the 50–400 μm vessels situated in the vasa nervorum, causing a length-dependent, asynchronous, and asymmetric pattern of neuropathy. The clinical findings of mononeuritis multiplex are numbness and weakness, which may occur either suddenly or subacutely.
Our patient presented with clinical features that suggested a symmetric small fiber neuropathy rather than a mononeuritis multiplex. Skin biopsy confirmed decreased small fiber density in the dermal and epidermal regions. Although our patient had glucose intolerance, her dysesthesias resolved quickly after she began receiving high doses of prednisone. A small fiber neuropathy due to impaired glucose tolerance would not have improved while receiving glucocorticoids.
The fact that some forms of systemic vasculitis cause a small fiber neuropathy is underappreciated. To our knowledge, this case represents the first time in which a skin biopsy sample was used to validate the clinical impression of small fiber neuropathy in a patient with MPA who did not have findings of disease in larger nerves. Lee et al (25) described skin biopsy findings supporting a small fiber neuropathy in 6 patients with concomitant mononeuritis multiplex due to nonsystemic vasculitic neuropathy. Lacomis et al (26) reported 2 patients with predominantly small fiber symptoms. In those patients, nerve biopsies showed a concomitant process leading to an axonal neuropathy. Prior descriptions of neuropathy in MPA have emphasized a subset of patients with a more indolent progression of symmetric polyneuropathies (27). It is likely that some of these patients have a concomitant small fiber neuropathy. Animal models of ischemic neuropathy have suggested that small-caliber nerves are more susceptible to ischemia than are larger, myelinated nerves (28).
Rheumatologists should appreciate both the utility and the limitations of electrodiagnostic studies in evaluating patients with potential vasculitic neuropathy. Electrodiagnostic studies are only helpful in detecting the presence of a neuropathy due to myelinated axons. In patients with mononeuritis multiplex, the finding of abnormal electrodiagnostic studies (e.g., decreased compound motor action potentials or fibrillation potentials in denervated muscles) can support the clinical impression of a large fiber neuropathy. In contrast, normal electrodiagnostic studies and sural nerve biopsies are consistent with clinical impressions of a small fiber neuropathy (29). Skin biopsy, a validated diagnostic tool in the evaluation of small fiber neuropathy (30), will constitute an increasingly important part of the evaluation of vasculitic neuropathies.
The lessons learned from this case include the fact that MPA can involve atypical organs; in this case, the bladder. IC/PBS is a diagnosis of exclusion, and in this patient, vasculitis of medium-sized arterioles and arteries led to symptoms of IC/PBS that defied a diagnosis for years. Some forms of systemic vasculitis can cause a small fiber neuropathy, the classic feature of which is pain. In small fiber neuropathy, both the neurologic examination and electrodiagnostic studies are normal. The diagnosis can be made by skin biopsy.
The patient enrolled in a randomized, double-blind, placebo-controlled trial of rituximab plus glucocorticoids versus cyclophosphamide plus glucocorticoids. She achieved disease remission quickly on the therapy to which she was assigned, tapered off prednisone after 5 months, and remains in complete remission 2 years after diagnosis. Ironically, her greatest clinical improvement followed her cystectomy, which was performed even before the diagnosis was made. She had no further abdominal symptoms following her cystectomy. The symptoms of her small fiber neuropathy resolved as her MPA came under control and she entered disease remission.
Dr. Stone had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study design. Birnbaum, Stone.
Acquisition of data. Birnbaum, Chai, Stone.
Analysis and interpretation of data. Birnbaum, Polydefkis, Stone.