|
|
| |||||||||||||||||||||||||||||||||
Background Idiopathic autonomic neuropathy is a severe, subacute disorder with a presumed autoimmune basis. It is indistinguishable from the subacute autonomic neuropathy that may accompany lung cancer or other tumors. Autoantibodies specific for nicotinic acetylcholine receptors in the autonomic ganglia are potentially pathogenic and may serve as serologic markers of various forms of autoimmune autonomic neuropathy.
Methods We tested serum from 157 patients with a variety of types of dysautonomia. Immunoprecipitation assays with iodine-125–labeled epibatidine and solubilized human neuroblastoma acetylcholine receptors were used to detect autoantibodies that bound to or blocked ganglionic receptors.
Results Ganglionic-receptor–binding antibodies were found in 19 of 46 patients with idiopathic or paraneoplastic autonomic neuropathy (41 percent), in 6 of 67 patients with postural tachycardia syndrome, idiopathic gastrointestinal dysmotility, or diabetic autonomic neuropathy (9 percent), and in none of 44 patients with other autonomic disorders. High levels of the binding antibodies correlated with more severe autonomic dysfunction (including the presence of tonic pupils). Levels of these antibodies decreased in patients who had clinical improvement. All seven patients with ganglionic-receptor–blocking antibodies had ganglionic-receptor–binding antibodies and had idiopathic or paraneoplastic autonomic neuropathy.
Conclusions Seropositivity for antibodies that bind to or block ganglionic acetylcholine receptors identifies patients with various forms of autoimmune autonomic neuropathy and distinguishes these disorders from other types of dysautonomia. The positive correlation between high levels of ganglionic-receptor antibodies and the severity of autonomic dysfunction suggests that the antibodies have a pathogenic role in these types of neuropathy.
Dysfunction of the autonomic nervous system is prominent in several neurologic disorders. Dysautonomia may occur in association with a degenerative disorder of the central nervous system, as in multiple-system atrophy (also known as the Shy–Drager syndrome),5 or may be due to dysfunction of the peripheral autonomic nervous system (comprising the autonomic nerves and ganglia). A slowly progressive disorder of the autonomic nerves, presumed to be degenerative, was first described by Bradbury and Eggleston6 and is now known as pure autonomic failure. Chronic acquired autonomic neuropathy may also occur in patients with amyloidosis or diabetes, and autonomic dysfunction is a feature of some inherited neuropathies.
In patients with idiopathic autonomic neuropathy, severe panautonomic failure develops in a subacute manner (over a period of days or weeks).7,8 The course of their disease is generally monophasic, with slow and incomplete recovery.8 Sympathetic failure is manifested as severe orthostatic hypotension and anhidrosis, and parasympathetic failure as dry mouth, sexual dysfunction, an impaired pupillary response to light and accommodation, and a fixed heart rate. Many patients have gastrointestinal dysmotility and present with anorexia, early satiety, abdominal pain and vomiting after eating, constipation, or diarrhea. Motor-nerve and sensory-nerve abnormalities are minimal or absent. An identical syndrome of subacute autonomic failure may occur as an indirect immunologic effect of cancer, especially small-cell carcinoma of the lung.9
The concept that idiopathic and paraneoplastic forms of autonomic neuropathy are immune-mediated (and thus may be described as forms of autoimmune autonomic neuropathy) is based in part on the associations of these syndromes with lung cancer,9 myasthenia gravis, and thymoma10,11 and on their clinical similarity to the acute autonomic disturbances that may accompany Guillain–Barré syndrome.12 The onset of idiopathic autonomic neuropathy frequently follows the resolution of a viral illness. The level of protein in the spinal fluid may be high, and perivascular lymphocytic infiltrates may be seen in biopsy specimens of the sural nerve.12 There have been individual case reports of benefit from intravenous immune globulin or plasma exchange.13,14,15 Our previous identification of ganglionic-receptor antibodies in five patients with idiopathic or paraneoplastic autonomic neuropathy supports the classification of these disorders as forms of autoimmune autonomic neuropathy.4
The clinical features of autoimmune autonomic neuropathy, with or without cancer, are usually distinct from those of degenerative autonomic disorders. For example, in patients with pure autonomic failure, the onset of orthostatic hypotension is usually insidious, and there is no prominent gastrointestinal dysmotility, dryness of the mouth, or pupillary dysfunction.12 Nevertheless, early differentiation among these disorders is sometimes difficult. The accuracy of the diagnosis has important prognostic and therapeutic implications, because pure autonomic failure is a slowly progressive disorder, whereas idiopathic or paraneoplastic autonomic neuropathy is a life-threatening disorder that may respond to immunomodulatory therapy or to therapy for cancer. In some cases of idiopathic autonomic neuropathy, the condition improves spontaneously.
Our preliminary observations suggested that antibodies that bind to ganglionic receptors may be useful diagnostic markers of various forms of autoimmune autonomic neuropathy and that they may be pathogenic.4 In the current study, which involved a large number of patients with a variety of autonomic disorders, we investigated whether antibodies that bind to or block ganglionic receptors are markers that may help to distinguish autoimmune from degenerative forms of dysautonomia. We measured the serum levels of ganglionic-receptor–binding antibodies and assessed ganglionic-receptor–blocking activity in these patients, and we analyzed the serologic findings in relation to the results of quantitative tests of autonomic function.
Methods
Patients
This study was approved by the institutional review board of the Mayo Clinic in Rochester, Minnesota, and did not require written informed consent. Serum samples from patients with autonomic dysfunction were submitted to the neuroimmunology laboratory of the clinic by the neurologists or gastroenterologists who were caring for these patients. Patients with a diagnosis of the Lambert–Eaton syndrome were excluded. To assign a clinical diagnosis, two or more investigators (at least one of whom was unaware of the results of the antibody assays) reviewed the patients' histories, notes from their physical examinations, and the results of autonomic-function tests, gastrointestinal-motility studies, and other clinical tests. Forty-seven patients were excluded from the study for one of the following reasons: a lack of objective evidence of autonomic impairment at the time of serologic testing (the most common reason for exclusion), insufficient information to establish the type of dysautonomia, or a lack of consensus among the reviewers with respect to the diagnosis.
A total of 157 patients were included in the study and were classified as having one of the following diagnoses: idiopathic autonomic neuropathy, paraneoplastic autonomic neuropathy, postural tachycardia syndrome, idiopathic gastrointestinal dysmotility, diabetic autonomic neuropathy, pure autonomic failure, multiple-system atrophy, or nondiabetic sensorimotor and autonomic neuropathy. The last category included Guillain–Barré syndrome, chronic inflammatory demyelinating polyradiculoneuropathy, hereditary sensory and autonomic neuropathy, and idiopathic small-fiber peripheral neuropathy with objective autonomic involvement. Fifty-one of the 157 patients had participated in our preliminary study of ganglionic-receptor–binding antibodies4 (9 of whom had idiopathic autonomic neuropathy, 3 paraneoplastic autonomic neuropathy, and 39 other dysautonomias).
Autonomic-Function Studies
A validated set of autonomic-function tests was used to assess the distribution and severity of sudomotor, cardiac vagal, and adrenergic failure.16 The Quantitative Sudomotor-Axon Reflex Test17 was used to evaluate postganglionic sympathetic function. Recordings were made from one site on an arm and three sites on a leg. Cardiac vagal function was evaluated on the basis of the heart-rate response to deep breathing and the Valsalva maneuver. Adrenergic function was evaluated on the basis of beat-to-beat responses in blood pressure during head-up tilting and the Valsalva maneuver. The results were compared with control values derived from studies of 357 healthy subjects 10 to 83 years old.17
The Composite Autonomic Severity Scale combines the results of these three sets of autonomic-function tests; sudomotor function is assigned a score of 0 to 3 points, cardiac vagal function 0 to 3 points, and adrenergic function 0 to 4 points (higher scores indicate a greater deficit). The semiquantitative composite score for autonomic failure thus ranges from a total value of 0 (normal function) to 10 (maximal deficit); a score of 7 or more indicates severe autonomic failure.18 The score is adjusted for the effects of sex and age.
The thermoregulatory sweat test was used to measure the amount and pattern of sweat output over the entire anterior surface of the body. Sweating was detected by the change in color of alizarin red, an indicator powder that was applied to the skin. The patient was then subjected to a standardized thermal stimulus and photographed; results were expressed as the percentage of body-surface area that did not sweat (100 percent indicates complete anhidrosis).19
Gastrointestinal-motility studies were performed in patients with symptoms of gastrointestinal dysmotility. These studies included scintigraphic testing of gastric and intestinal transit and endoscopy with esophagogastroduodenal manometry to investigate the patterns of contractions before and after food intake.
Antibody Assays
To test serum for antibodies that bind to neuronal ganglionic receptors,4 antigen was solubilized from a human neuroblastoma cell line (IMR-32) and bound to epibatidine (final concentration, 2 nmol per liter),20 a high-affinity agonist, which was radiolabeled with iodine-125. Five microliters of serum was added to each of two identical 40-fmol aliquots of the 125I-epibatidine–ganglionic-receptor complex. After incubation for 16 hours at 4°C, human IgG and IgM were precipitated by the addition of goat antiserum, and radioactivity was then measured (in terms of counts per minute) in washed pellets that were obtained by repeated centrifugation. Positive and negative control samples of human serum were included in each assay. Serum samples that yielded precipitates with more radioactivity than the mean value +1 SD obtained for four normal control samples run in the same assay were retested at least once and were diluted 10-fold if needed to attain a linear range of antigen binding. The antigen-binding capacity of each serum sample was expressed in terms of picomoles of 125I-epibatidine–ganglionic-receptor complex precipitated per liter of serum. Studies to date have validated a value of 50 pmol per liter as the upper limit of normal for serum from healthy, nonimmunosuppressed subjects.
To detect antibodies that block the ganglionic receptor (i.e., that bind to or near to the agonist-binding site on the receptor), we modified the assay established for detecting antibodies that block the binding of 
-bungarotoxin to skeletal-muscle acetylcholine receptors.21 Twenty-five microliters of serum was added to 80-fmol aliquots of solubilized ganglionic receptor, and a total volume of 175 µl was incubated for 16 hours at 4°C before the addition of 125I-epibatidine (to a final concentration of 2 nmol per liter). After a 15-minute incubation at 22°C, the receptors were precipitated by sequential addition of a standard rat monoclonal antibody (Mab35, which binds to acetylcholine receptors)4 and goat antiserum that reacts with rat and human immunoglobulins. The mean radioactivity of precipitates obtained when acetylcholine receptor was incubated with normal human serum was defined as 100 percent binding of ligand (i.e., 0 percent blockade). Antibodies that interfere with the binding of 125I-epibatidine to ganglionic receptors decrease the amount of radioactivity precipitated. Results were expressed as the percentage blockade of ligand binding; the normal value established in studies of healthy subjects is less than 25 percent blockade.
Paraneoplastic autoantibodies specific for neuronal nuclear and cytoplasmic antigens were detected with a standardized indirect immunofluorescence assay.22
Results
Frequency of Ganglionic-Receptor Antibodies in Patients with Dysautonomia
Ganglionic-receptor antibodies were not detected in 133 healthy control subjects or in 13 control patients who were referred because they had gastrointestinal symptoms and who had normal results on motility studies. The antibody assays had good reliability (coefficient of variation, 23.0 with samples stored and tested repeatedly for six months), and the results were consistent when successive serum samples from individual subjects were tested.
Ganglionic-receptor–binding antibodies were detected in the serum samples from 25 of the 157 patients with various autonomic disorders (Table 1 and Table 2). Nineteen of these 25 patients (76 percent) had a clinical diagnosis of either idiopathic or paraneoplastic autonomic neuropathy. Of the five seropositive patients with paraneoplastic autonomic neuropathy, one had small-cell lung carcinoma, two had thymoma, one had bladder carcinoma, and one had rectal carcinoma. The 14 patients without evidence of cancer constituted 50 percent of the 28 patients classified as having idiopathic autonomic neuropathy.
|
|
|
Correlation of Levels of Ganglionic-Receptor–Binding Antibody with the Severity of Autonomic Dysfunction
Among the 19 patients with dysautonomia who were seropositive for ganglionic-receptor–binding antibody and who underwent detailed quantitative autonomic testing, high levels of antibodies were significantly correlated with more severe autonomic failure, as assessed by the score on the Composite Autonomic Severity Scale (Figure 2). Tonic pupils (an impaired pupillary response to light and to accommodation) were observed in 11 patients (8 with idiopathic and 3 with paraneoplastic autonomic neuropathy), 9 of whom had ganglionic-receptor–binding antibodies.
|
|
|
Five of the 18 patients with paraneoplastic autonomic neuropathy were seropositive for ganglionic-receptor–binding antibody. Twelve of the 18 patients had small-cell lung carcinoma, and 10 of these 12, and no others, had type 1 antineuronal nuclear autoantibody (also called anti-Hu). Type 1 antineuronal nuclear autoantibody has been reported to be a marker of paraneoplastic autonomic neuropathy and gastrointestinal dysmotility related to small-cell lung carcinoma.23,24,25
Discussion
The results of this study justify the classification of idiopathic and paraneoplastic forms of autonomic neuropathy as autoimmune autonomic neuropathy. Ganglionic-receptor autoantibodies with specific binding or blocking actions were highly associated with both of these diagnoses, and positive results on serologic tests distinguished these disorders from degenerative forms of dysautonomia. High levels of ganglionic-receptor–binding antibody correlated significantly with the severity of autonomic dysfunction, and the level of such antibodies decreased or became undetectable with improvement of autonomic function. These findings suggest that the autoantibodies contribute to the pathogenesis of subacute autonomic failure. Although seropositivity for the ganglionic-receptor–binding or ganglionic-receptor–blocking antibodies unambiguously identified patients with autoimmune autonomic neuropathy, a negative result did not rule out this diagnosis.
Clinical features that are characteristic of autoimmune autonomic neuropathy are a subacute onset, prominent symptoms of gastrointestinal dysmotility, and an abnormal pupillary response to light and to accommodation. An impaired pupillary response in patients with severe parasympathetic dysfunction was the clinical feature that most reliably predicted seropositivity for the ganglionic-receptor–binding antibody. This observation is consistent with the striking mydriasis reported in mutant mice lacking the 3 subunit of the ganglionic acetylcholine receptor.26
Because the various forms of acquired dysautonomia may be difficult to distinguish on the basis of the clinical history and clinical examination alone, ancillary testing remains important for establishing a diagnosis of autoimmune autonomic neuropathy. For example, a case of isolated gastrointestinal dysmotility would be difficult to distinguish from autonomic neuropathy without the results of specific laboratory tests of cardiovascular and sudomotor autonomic functions. Serologic testing for ganglionic-receptor–binding antibodies early in the clinical evaluation of patients with autonomic disorders should improve diagnostic accuracy and may help identify patients who are likely to benefit from immunomodulatory therapy. The clinical characteristics of the seropositive patients with autonomic neuropathy in our study were indistinguishable from those of patients with autonomic neuropathy whose symptoms were reported to have improved in response to intravenous immune globulin therapy,13,14 probably as a result of accelerated catabolism of pathogenic IgG.27
Six of the patients in this study who were seropositive for the ganglionic-receptor–binding antibody did not have a clinical diagnosis of idiopathic or paraneoplastic autonomic neuropathy. Of these six patients, three had a diagnosis of idiopathic gastrointestinal dysmotility, two had diabetic autonomic neuropathy, and one had postural tachycardia syndrome. Their serum levels of ganglionic-receptor–binding antibodies were lower than those in most of the other seropositive patients. Formal testing of cardiovascular and sudomotor autonomic function was not performed in three of them (two with idiopathic gastrointestinal dysmotility and one with diabetic autonomic neuropathy). Thus, signs of a more diffuse but mild dysautonomia may not have been recognized because of the prominent gastrointestinal symptoms and signs.28 The fact that these patients were seropositive for the ganglionic-receptor–binding antibody supports the concept that in some cases, isolated gastrointestinal dysmotility, postural tachycardia syndrome, and autonomic neuropathy in diabetic patients represents variant presentations of autoimmune autonomic neuropathy.28,29,30 In our study, there were no overt clinical differences between seropositive patients and seronegative patients with these diagnoses.
The sensitivity of the assay for ganglionic-receptor–binding antibody in patients with idiopathic autonomic neuropathy (50 percent of whom were seropositive) is lower than the seropositivity rates of more than 90 percent for assays of the muscle-receptor–binding antibody or modulating antibody in adult patients with generalized myasthenia gravis31 and for the assay of the P/Q-type calcium-channel–binding antibody in patients with the Lambert–Eaton syndrome.32 However, unlike autoimmune disorders of the neuromuscular junction, autoimmune autonomic neuropathy is often a monophasic illness. For many of the patients in this study, serologic testing was not performed until several months after the onset of symptoms. Thus, transient seropositivity during the acute phase of the illness may have been missed. Another possibility is that the neuronal antigens that are the targets of pathogenic autoantibodies may differ among individual patients with autonomic neuropathy. This appears to be the case for patients with autoimmune disorders of neuromuscular hyperexcitability, in whom both voltage-gated potassium channels and prejunctional neuronal nicotinic acetylcholine receptors have been implicated as autoantigens.33,34 It is also possible that nonimmune mechanisms account for some of the seronegative findings in patients with subacute autonomic failure.
In addition to suggesting a useful serologic marker for establishing a diagnosis of autoimmune autonomic neuropathy, our data indicate that ganglionic-receptor antibodies may be effectors of autonomic dysfunction in patients with idiopathic or paraneoplastic autonomic neuropathy. First, we documented a significant correlation between the serum level of ganglionic-receptor–binding antibody and the severity of autonomic dysfunction. Second, decreases in the level of this antibody were accompanied by improvement in clinical and laboratory measures of autonomic function. Third, ganglionic-receptor antibodies with blocking specificity, which have the greatest potential to impair ganglionic synaptic transmission in vivo, disappeared from the serum in parallel with clinical improvement. Similarly, in patients with myasthenia gravis, muscle-receptor–blocking antibodies are less prevalent than muscle-receptor–binding antibodies,35 and the levels of both antibodies decrease with clinical improvement.36 Currently, we have only indirect evidence that ganglionic-receptor antibodies disrupt autonomic function. Definitive proof will require demonstration that an IgG antibody that is specific for the ganglionic receptor interferes with synaptic transmission when administered to laboratory animals or applied to autonomic ganglia in vitro.
Supported by the Mayo Foundation and the Mayo Clinic Cancer Center.
We are indebted to members of the Mayo Clinic departments of neurology and gastroenterology for identifying patients for this study and to Thomas Kryzer, Mary Lohse, Evelyn Posthumus, and Tonette Opfer-Gehrking for providing excellent technical support.
Source Information
From the Departments of Neurology (S.V., P.A.L., R.D.F., V.A.L.), Immunology (V.A.L.), and Laboratory Medicine and Pathology (V.A.L.) and the Division of Gastroenterology and Hepatology (G.F.), Mayo Clinic, Rochester, Minn.; and the Department of Neurology and Neurosurgery, Montreal Neurological Hospital and Institute, McGill University, Montreal (J.D.S.).
Address reprint requests to Dr. Vernino at the Department of Neurology, Mayo Clinic, 200 First St., S.W., Rochester, MN 55905, or at verns{at}mayo.edu.
References
3 neuronal nicotinic acetylcholine receptor. Proc Natl Acad Sci U S A 1999;96:5746-5751.
| |||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
