559

severity and distribution in various diseases.2 Direct evidence of a pathogenetic role for ANCA has come from the demonstration in vitro that ANCA autoantibodies induce neutrophil degranulation with production of oxygen radicals.3ANCA are present in uveitis and retinal vasculitis.4 In our patient, immunoglobulin replacement therapy seemed to result in intermittent passive administration of cANCA, with induction of retinal vasculitis and uveitis. Focal infarction by secondary immune processes dependent on ANCA in the choriocapillaries could account for the patchy pale fundal lesions, and damage to retinal vessels would alter the blood-retinal barrier, allowing inflammatory cells to enter the vitreous. Although a case of XLA associated with uveitis has been recorded5 our report is, as far as we are aware, the first association of retinal vasculitis and fundal changes with uveitis with XLA. Manchester Royal Eye Hospital, Manchester M13 9PH, UK, Department of Immunology,

Hope Hospital, Salford; and Duchess of York Children’s Manchester

Hospital,

W. AYLIFFE M. HAENEY S. C. ROBERTS M. LAVIN

1. Wilk A. Delineation of a standard procedure for indirect immunofluorescence detection of ANCA. APMIS 1989; 97 (suppl 6): 12. 2. Pnem ILA, Oosterhuis JA. Birdshot chorioretinopathy: clinical charactenstics and evolution. Br J Ophthalmol 1989, 72: 646-59. Jennette JC, Wilkman AS, Falk RJ. Antineutrophil cytoplasmic antibody-associated glomerulonephritis and vasculitis. Am J Pathol 1989; 135: 921-30 4. Falk RJ, Terrell RS, Charles LA, Jennette JC. Antineutrophil cytoplasmic antibodies induce neutrophils to degranulate and produce oxygen radicals in vitro. Proc Natl 3.

Acad Sci USA 1990; 87: 4115-19. CM, Buckley RII, Mohanakumar T, et al. Persistent and fatal centralnervous-system echovirus infections m patients with agammaglobulinaemia. N Engl J Med 1977; 296: 1485-89

5. Wilfret

Type II collagen antibodies in patients with sensorineural hearing loss

Red-free fundus photograph of right eye (upper) and mid-phase fluorescein angiogram (lower).

Upper figure shows large peripapillary placoid-like lesion with multiple discrete smaller creamy white lesions scattered throughout posterior pole. Lower shows disc hyperfluorescence and focal staining of retinal vessels, small punctate hyperfluorescent outer retinal lesions are

evenly distributed

to

equator

Systemic prednisolone 60 mg daily led to resolution of the cellular infiltrate, and acuity improved to 6/6 in both eyes. Within 3 days of the next replacement infusion the vasculitis, panuveitis, severe vitreous infiltration, and decreased vision recurred. This responded only partly to systemic prednisolone (60 mg daily), but vasculitis disappeared with oral cyclosporin (12.5mg/kg daily) for 3 weeks. Although mild symptomless anterior uveitis persisted, the punctate lesions faded to very subtle depigmented areas and the placoid-like lesion resolved, leaving a pronounced depigmented area. Because of

vasculitis, serum samples were tested immediately before and after infusions of immunoglobulin for antibodies to neutrophil cytoplasmic antigens (cANCA), and were negative before infusion but became positive (titre 1/32) 2 h post infusion with a standardised technique.1 Several immunoglobulin preparations were repeatedly positive for cANCA reactivity at recommended concentrations and diluted to 10 g IgG/l. Immunoglobulin treatment was changed to ’Endobulin’ (Immuno) 0-2 g/kg intravenously every month since this preparation was negative for cANCA reactivity. So far, there has not been a recurrence of the posterior segment disease, although mild anterior uveitis persists and is controlled with twice daily

topical prednisolone. The ophthalmic signs

are best accounted for by underlying vasculitis. ANCA are present in patients with systemic vasculitis, and ANCA titre, pattern, and isotope have been linked to disease

SIR,-Bilateral progressive sensorineural hearing loss (BPSHL) is thought to have an autoimmune aetiology in a subset of patients.1 Helfgott and colleagues2 reported that the presence of antibodies to type II collagen might be predictive of corticosteroid responsiveness in BPSHL patients. However, the ELISA test for detection of type II collagen antibodies is in question because of the possibility of artifactual detection of immune complexes and non-collagen-specific immunoglobulins bound to the pepsinogen contaminant in the collagen preparation.3,4 Furthermore, the hydrophobicity of the Fc fraction of IgG may cause non-specific binding to the free active sites of the ELISA wells.3 We have reservations about Helfgott and colleagues’ conclusions because their ELISA method does not use a blocking reagent, and the reactivity of serum samples positive for type II collagen antibodies is not confirmed by other methods. The potential difficulties of artifacts can be avoided by improving the assay with heterologous serum samples and confirming the binding of type II collagen antibodies on western blots. Heterologous serum samples have been used by others to block remaining active binding sites of the ELISA wells in this assay.5,6 Additionally, heterologous immunoglobulins in the blocking agent and dilution buffer might also inhibit non-specific interactions with the pepsinogen contaminant. With 5% horse serum to block non-specific binding sites, we set up ELISAs for the detection of antibodies to both bovine type II collagen (Sigma) and chicken type II collagen (Genzyme) in disease and normal populations, defining the cut-off value as the mean optical density (+3 SD) obtained from normal sera. Antibodies reacting with bovine type II collagen were detected in one of eight patients with steroid-responsive inner ear disease, none of ten with Meniere syndrome, one of thirteen with sudden deafness, one of two with otosclerosis, and one of forty rheumatoid-factor-positive sera, none of twenty antinuclear-antibody-positive sera, and three of one hundred and three normal sera. The highest binding was seen in a patient with otosclerosis and the presence of specific bovine type II collagen antibodies was confirmed by western blot. With chicken type II collagen as antigen, autoantibodies to type II collagen were not detected in any of the eight patients with steroid-responsive inner-ear disease and ten with Meniere syndrome. However, antibodies reacting with chicken type II

560

collagen

were

detected in

one

of thirteen

patients with sudden

deafness and two of ninety-one healthy individuals. The low positivity of type II collagen antibodies in patients with inner ear disease is not consistent with an autoimmune response against a1(II) chains of type II collagen, which have a molecular mass of about 95 kDa. Additionally, the 68 kDa antigen of the inner ear might be more important in autoimmune sensorineural hearing loss.7 The discrepancies between our results and those of Helfgott et al might represent differences in patient populations, and they point to a need for the collaborative exchange of serum samples between laboratories investigating the cause of sensorineural hearing loss. Specialty Laboratories Inc. Santa Monica, California 90404, USA

Department of Neurology,

MADE SUTJITA JAMES B. PETER ROBERT W. BALOH

UCLA School of Medicine, Los Angeles

JOHN G. OAS

Department of Otolaryngology, Central Hospital, Falun, Sweden

CLAUDE LAURENT LEIF NORDANG

1. Hughes GB, Kinney SE, Bama BP, Calabese LH. Practical versus theoretical management of autoimmune inner ear disease. Laryngoscope 1984; 94: 758-67. 2. Helfgott SM, Mosciscki RA, San Martin J, et al. Correlation between antibodies to type II collagen and treatment outcome in bilateral progressive sensorineural hearing loss. Lancet 1991; 337: 387-89. 3. Kirk AP, O’Hara BP, Mageed RAK, et al. Pepsinogen: an immunoglobulin binding artifact in ’collagen’ preparations, Clin Exp Immunol 1986; 65: 671-78. 4. O’Hara BP, Pyle J, McCarthy D, Archer JR. Binding of monomeric and aggregated immunoglobulins to enzymes: a source of artifact in antibody assay. J Immunol Methods 1989, 116: 175-79. 5. Fujii K, Tsuji M, Murota K, et al. An improved enzyme-linked immunosorbent assay of anti-collagen antibodies in human serum. JImmunol Methods 1989; 124: 63-70. 6. Terato K, Shimozuru Y, Katayama K, et al. Specificity of antibodies to type II collagen in rheumatoid arthritis. Arthritis Rheum 1990; 33: 1493-500. 7. Harris JP, Sharp PA. Inner ear autoantibodies m patients with rapidly progressive sensorineural hearing loss. Laryngoscope 1990; 100: 516-24.

causes less inflammation and less hearing loss than that due to the other groups. This finding may arise because the immune system reacts less to the group B capsular polysaccharide, which is immunochemically identical to a fetal brain glycopeptide.1 This is also shown by the low immunogenicity of this polysaccharide, which hinders the development of an effective vaccine against N meningitidis serogroup B. Other sequelae were evenly distributed among the serogroups. Why are unusual meningococcal serogroups associated with hearing loss more frequently (24% in our study) than common serogroups? There may be immunological reasons, since many patients with disease due to the uncommon groups have a complement deficiency.s

group B disease

Netherlands Reference Laboratory for Bacterial Meningitis of the National Institute of Public Health and Environmental Protection, University of Amsterdam, 1105 AZ Amsterdam, Netherlands 1.

L. SPANJAARD P. BOL S. DE MARIE H. C. ZANEN

Spanjaard L, Bol P, De Marie S, Zanen HC. Association of meningococcal serogroups with the course of disease in the Netherlands, 1959-83. Bull World Health Organ

1987; 65: 861-68. 2. Netherlands Reference Laboratory for Bacterial Meningitis (RIVM/UvA). Bacterial meningitis in the Netherlands; annual report 1990. Amsterdam: University of Amsterdam, 1991. 3. Spanjaard L. Epidemiology of bacterial meningitis in the Netherlands; volume 1. Chapter 8. Thesis. Amsterdam: University of Amsterdam, 1986. 4. Finne J, Leinonen M, Mäkelä PH. Antigenic similarities between brain components and bactena causing meningitis; implications for vaccine development and pathogenesis. Lancet 1983; ii: 355-57. 5 Fijen CAP, Kuijper ED, Hannema AD, Sjoholm AG, Van Putten JPM. Complement deficiencies m patients over ten years old with meningococcal disease due to uncommon serogroups. Lancet 1989, n: 585-88.

&bgr;-amyloid protein immunoreactivity in patients with inclusion-body

muscle of

myositis Deafness after

SIR,—Inclusion-body myositis(1HM)

meningococcal meningitis

SIR,-Dr Mayatepek and colleagues (Nov 23, p 1331) reported a

prevalence of deafness after meningitis due to Neisseria meningitidis serogroup B (33%) than when unusual serogroups were cultured (20%). Dr Moss (Dec 21/28, p 1602), however, did not find a significant difference in the frequency of deafness between children with meningococcal meningitis (5 % ) and controls (3%). We report here the findings of a retrospective study.’1 Since 1959, strains and samples of cerebrospinal fluid (CSF) from patients with meningococcal disease in the Netherlands have been collected and investigated at the Netherlands Reference Laboratory for Bacterial MeningitisWe examined the medical histories of 1221 patients (1959-83).’ 62 patients (5%) died and 91 (8%) of the 1159 survivors had sequelae, 37 (3-2%) of whom had loss of hearing (table). Loss of hearing was seen less often in group B disease (14/744, 2%; 95% confidence interval [CI] 1.0-3.1 1 %) than in all other groups (23/415; 5-5%; 95% CI 35--82%). CSF values (day of admission) in patients with group B disease were consistent with a lesser degree of inflammation (eg, lower white cell count and protein content) than were those with group A or C disease, whereas lower

the duration of symptoms before admission was much the

same

in

all groups3.

Although in this study the proportion of patients with loss of hearing was low (probably because of missing data) there is no reason to

suppose that the serogroup distribution differed among

patients whose data

were not

available. Our results indicate that

SEQUELAE AFTER MENINGOCOCCAL DISEASE IN THE NETHERLANDS (1959-83) ACCORDING TO SEROGROUP

*X

(n=3),(9).Z(1).), non-groupable (4)

is a

usually sporadic

progressive myopathy of unknown cause and pathogenesis.’ It has clinical and pathological similarities to polymyositis, but is either not responsive, or only moderately so, to immunosuppressive treatment. Light-microscopic features of muscle-biopsy specimens include mononuclear inflammatory cells varying from abundant to none, atrophic muscle fibres, and muscle fibres with rimmed vacuoles that contain characteristic cytoplasmic tubulofilaments (CTFs) identified by electronmicroscopy.2We have shown that vacuolated muscle fibres contain strong ubiquitin immunoreactivity, which by immunoelectronmicroscopy was localised to CTFs.3 Vacuoles also show Congo-red positivity, indicating amyloid,4 but the type of amyloid protein has not been identified. We report an investigation of 10 patients (aged 42-74, median 64) with inclusion-body myositis, including 1 hereditary case, and 14 control patients (aged 5-79, median 60), including 7 with polymyositis, 1 Duchenne muscular dystrophy, 4 amyotrophic lateral sclerosis, and 2 normal muscle. Vacuolated muscle fibres contained strong accumulation of &bgr;-amyloid protein (&bgr;-AP) within the vacuoles and sometimes in vacuole-free cytoplasm (figure). &bgr;-AP was identified by immunocytochemical staining with a well-characterised monoclonal antibody G-OP-1directed against sequence 8-17 of a synthetic &bgr;-AP.5 Omission of the primary antibody or replacing it with non-immune serum resulted in negative staining. With light microscopy, &bgr;-AP immunoreactivity closely co-localised with ubiquitin immunoreactivity. The amyloid inclusions were crystal-violet positive (metachromatic red) in all patients with sporadic disease, but not in the patient with hereditary disease. None of the muscle biopsy specimens from controls had &bgr;-AP-positive inclusions characteristic of inclusion-body myositis. &bgr;-AP was discovered in and first sequenced from the amyloid fibrils in blood-vessels of patients with Alzheimer’s diseaseand it has received considerable attention in the pathogenesis of this disease.’ In brain of patients with Alzheimer’s disease, &bgr;-AP immunoreactivity is present both in the amyloid fibrils (presumably &bgr;-pleated sheets), and located in blood-vessels and cores of senile plaques, as well as in non-fibrillar Congo-red-negative diffuse plaques, whereas ubiquitin immunoreactivity is seen at both locations.

Type II collagen antibodies in patients with sensorineural hearing loss.

559 severity and distribution in various diseases.2 Direct evidence of a pathogenetic role for ANCA has come from the demonstration in vitro that ANC...
492KB Sizes 0 Downloads 0 Views