ORIGINAL ARTICLE
High prevalence of inclusion body myositis in Norway; a population-based clinical epidemiology study €rnee, L. Grøvlef, J. T. Grana G. C. Doblouga, E. A. Antalb, L. Svebergc, T. Garena, H. Bitterd, J. Stja a,g and Ø. Molberg
EUROPEAN JOURNAL OF NEUROLOGY
a Department of Rheumatology, Oslo University Hospital (OUH), Oslo; bDepartment of Pathology, OUH, Oslo; cDepartment of Neurology, OUH, Oslo; dDepartment of Rheumatology, Sørlandet Hospital, Kristiansand; eDepartment of Rheumatology, Betanien Hospital, Skien; fDepartment of Rheumatology, Sykehuset Østfold, Moss; and gInstitute of Clinical Medicine, University of Oslo, Oslo, Norway
Keywords:
epidemiology, European Neuro-Muscular Centre Criteria, idiopathic inflammatory myopathy, sporadic inclusion body myositis Received 2 September 2014 Accepted 17 October 2014 European Journal of Neurology 2015, 22: 672–680 doi:10.1111/ene.12627
Background and purpose: Knowledge about the occurrence of sporadic inclusion body myositis (sIBM) in the general population is limited. Here, our aim was to identify and characterize every sIBM patient living in southeast Norway (population 2.64 million) from 2003 to 2012. Method: Two sIBM case finding strategies were applied. First, all hospital databases in southeast Norway were screened to identify cases with sIBM-compatible International Classification of Diseases 10 (ICD-10) codes. These cases were then manually chart reviewed. Secondly, all muscle histology reports encoded with inflammation were independently reviewed. Finally, cases were classified according to the 1997 and the 2011 European Neuro-Muscular Centre (ENMC) Research Diagnostic Criteria for sIBM. Results: The combined case finding strategy identified 3160 patients with sIBM compatible ICD-10 codes, and a largely overlapping cohort of 500 patients having muscle biopsies encoded with inflammation. Detailed retrospective review of chart and histology data showed that 95 patients met the 2011 ENMC sIBM criteria and 92 met the 1997 criteria. Estimated point prevalence of sIBM was 33/1 000 000, equal with both criteria sets. Mean age at diagnosis was 66.9 years and mean diagnostic delay was 5.6 years. Chart review revealed higher frequencies of dysphagia (94% vs. 65%) and anti-Sjøgren syndrome A antibodies (39% vs. 12%) in female sIBM patients (n = 40) than in males. Coexisting rheumatic diseases were present in 25% of sIBM cases, with Sjøgren’s syndrome in 10%. Conclusion: An estimated point prevalence of sIBM seven times higher than previously observed in Europe is reported. Our data show considerable diagnostic delay, a major challenge with new sIBM treatments in the pipeline.
Introduction Sporadic inclusion body myositis (sIBM) was first acknowledged as an entity separate from polymyositis (PM) in 1971 [1], but descriptions of histopathological and clinical features compatible with sIBM exist from the mid-1960s [2,3]. Traditionally, sIBM has been Correspondence: G. C. Dobloug, Department of Rheumatology, Oslo University Hospital, Postbox 4950, Nydalen, N-0424 Oslo, Norway (tel.: +47 23074878; fax: +47 22844651; e-mails: [email protected] or [email protected]).
672
regarded as one of the three idiopathic inflammatory myopathies (IIM), along with polymyositis (PM) and dermatomyositis (DM), and it may account for 30% of IIM cases [4,5]. There is growing evidence, however, that the disease mechanisms in sIBM differ from PM and DM [4] and represent a distinct separate entity with both inflammatory and degenerative changes [6–9]. Recent large-sized clinical studies provide further support to the notion that the muscle disease pattern in sIBM is highly characteristic, with asymmetric distribution, slow progression of proximal and distal
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
weakness and predominant involvement of the quadriceps and finger flexor muscles [10,11]. Dysphagia appears to be very frequent and can be disabling and potentially life-threatening [10,12,13]. Unlike PM and DM, sIBM is unresponsive to immune-modulating treatments. Hence, even though involvement of other organs than muscles is rare, sIBM has an end-stage of major disabilities [11,14]. There is no doubt that the light microscopy findings in sIBM muscle may be highly characteristic with coexisting processes suggestive of inflammation (endomysial infiltrates and lymphocyte invasion in morphologically normal myocytes) and degeneration (eosinophilic cytoplasmic inclusions and rimmed vacuoles) [15], but it is important to note that any given sIBM biopsy may lack any of these characteristic microscopic features [15]. Since the cause or pathogenesis of IBM is not known, there is no ‘gold standard’ for the diagnosis. There is an ongoing debate on how sIBM should optimally be diagnosed. The first widely used diagnostic criteria for sIBM, suggested by Griggs et al. in 1995 [16], were primarily based on histology findings. In 1997, the European Neuro-Muscular Centre (ENMC) produced a new criteria set which allowed for sIBM diagnosis in cases with highly suggestive clinical features but incomplete histology [17]. Diagnostic criteria primarily based on clinical findings were first proposed by Benveniste and Hilton-Jones in 2009 [18]. These criteria only required that histology was supportive and not inconsistent with sIBM [18]. The new ENMC diagnostic criteria (2011 ENMC IBM Research Diagnostic Criteria) [19] also emphasize clinical phenotype rather than pathology and additionally aim to allow for earlier diagnosis than previous criteria sets. Few studies on the epidemiology of sIBM have been performed. The first, large chart review study on sIBM was a nationwide collaborative cross-sectional study performed in the Netherlands in 1999 [20]. The study applied the 1997 ENMC criteria and identified 76 patients with sIBM, giving an estimated population prevalence of 4.9/1 000 000 [20]. The authors reported that prevalence was probably underestimated, but to date there are no equivalent European studies for comparison. Three Australian surveys, all based on local histopathology criteria, reported sIBM prevalence ranging from 9.3/1 000 000 to 50.5/1 000 000 [21–23]. In Japan, the prevalence of sIBM judged by the combination of several sIBM criteria sets [15,16,18] was estimated at 9.8/1 000 000 in 2003, with an increasing prevalence over the last decade [24,25]. In contrast, a recent biopsy-register study from Turkey reported only one sIBM case per million, possibly reflecting differences in sIBM prevalence across genetic backgrounds [26].
© 2014 EAN
673
Data on the occurrence of sIBM in the general population are limited. In the current study, our aim was to identify all sIBM cases in southeast Norway (with a denominator population of 2.64 million) and describe the clinical characteristics, laboratory findings and imaging data in the resulting cohort by retrospective chart and histology report review.
Materials and methods Study cohort
The sIBM study was part of a larger epidemiology study aiming to identify all IIM patients in southeast Norway. The PM/DM part of this IIM study was recently published [27]. Southeast Norway consists of 10 counties with 2 642 246 inhabitants and includes the largest cities in Norway. The main hospital in the region is Oslo University Hospital (OUH), which is the primary hospital for the city of Oslo (with 600 000 inhabitants) and the referral hospital for all the other hospitals in southeast Norway. In Norway, patients with IIM are followed by specialists at public hospitals, where sIBM patients are cared for either by neurologists or by rheumatologists. Patients included were residential and seen at hospitals in the health region between January 2003 and December 2012, a 10-year period. Study inclusion criteria
Patients were included if they fulfilled the following criteria: (i) disease classifiable as sIBM by the 1997 ENMC sIBM criteria and/or the 2011 ENMC IBM Research Diagnostic Criteria [17,19]; (ii) exclusion of PM or DM as possible diagnoses; (iii) inclusion body myositis not explained by familial disease; (iv) patient registered in the Norwegian Central Population Register with a home address in southeast Norway between 1 January 2003 and 31 December 2012. Case finding strategy
Two major acquisition routes were utilized to identify all the adult sIBM patients: (i) chart review of all potential IIM cases and (ii) retrospective review (performed by a neuropathologist) of all muscle biopsy reports encoded with inflammation. First, all potential IIM cases were identified by searches across patient administrative databases. Initially, the database at OUH was screened using the following International Classification of Diseases 10 (ICD-10) codes: M33.1 (dermatomyositis), M33.2 (polymyositis), M33.9 (unspecified PM/DM), M60.1
674
G. C. DOBLOUG ET AL.
(interstitial myositis), M60.8 (specified myositis), M60.9 (unspecified myositis), G72.4 (inflammatory myopathy, not classified elsewhere), G72.8 (other specified myopathies), G72.9 (unspecified myopathy), G73.7 (myopathy associated with disease classified elsewhere) [27]. Preliminary data from OUH showed that no sIBM cases were G-coded. Hence, the searches undertaken at the other southeast Norway hospitals were limited to the six M33 and M60 codes. The charts of all the patients identified were manually reviewed by the principal investigator (CD). The muscle biopsies taken in southeast Norway between 2003 and 2012 were examined by five neuropathologists at two laboratories in Oslo (Rikshospitalet and Ulleval) and one in Tromsø. From 2010, the two laboratories in Oslo were merged within the Department of Pathology, OUH. For our review, all the muscle histology reports encoded with inflammation in the Systematic Nomenclature of Medicine (SNOMED) code system were reviewed by a neuropathologist (EAA), and the following parameters were recorded; endomysial inflammatory infiltrates, lymphocyte invasion in structurally normal myocytes, rimmed vacuoles, MHC 1 expression (any and general) and the presence of inclusion body filaments at ultra-structural examination. The neuropathologist had access to muscle biopsy referral information and the pathology reports produced by the five other neuropathologists, but not the clinical charts. Finally, all the patients identified by chart review and/or muscle biopsy review were discussed on a case to case notion and scored, primarily according to the 1997 sIBM ENMC criteria and later by the 2011 ENMC IBM Research Diagnostic Criteria. Recording of patient data and items assessed by the 1997 and 2011 ENMC criteria
Predefined registration forms were used to record hospital chart data. Time of symptom onset and total disease duration, defined as the time from diagnosis to study end, or the time of death were recorded. Scoring of the clinical and histopathology items according to the 1997 and 2011 ENMC criteria was performed as follows: (i) proximal muscle weakness, described by a specialist (rheumatologist or neurologist) during clinical examination as weakness (and most often also atrophy) involving thigh and/or shoulder/neck muscles weakness was quantified by the Medical Research Council (MRC) scale (0–5) and/or manual muscle testing (MMT) performed by physiotherapists; (ii) distal muscle weakness, described by a specialist during clinical examination as weakness in the finger flexors and/or the combined presence of reduced grip strength
and atrophy of ulnar forearm muscles distal weakness was not quantified but was registered as present or absent; (iii) knee extension weakness, described by a specialist and quantified by MRC and/or MMT; (iv) slowly progressive course, chart description of weakness progressing slowly over years; (v) sporadic disease, no chart information on familial clustering; (vi) total disease duration over 12 months; (vii) age of onset above 30 years for 1997 criteria and above 45 for 2011 criteria; (viii) maximum creatine kinase (CK) levels, recorded from chart data; (ix) muscle biopsy parameters recorded from pathology reports: endomysial inflammatory infiltrates, lymphocyte invasion in structurally normal myocytes, rimmed vacuoles, MHC 1 expression (any and general), the presence of inclusion body filaments at ultra-structural examination. Other clinical information recorded from patient charts
Myalgia, arthritis (as defined by a doctor), arthralgia, dysphagia, sicca symptoms, Raynaud phenomenon, cough, dyspnoea, erythrocyte sedimentation rate (ESR), anti-nuclear antibodies (ANA), anti-Sjøgren syndrome A (SSA) antibodies, myositis specific antibodies (anti-Jo-1, PL-7, PL-12, SRP and Mi-2), electromyography/neurography, magnetic resonance imaging (MRI) of muscle, dynamic X-ray of the oesophagus and high resolution chest computed tomography were recorded. Occurrence of chronic heart and/or lung disease was also recorded. Time and cause of death was recorded when applicable. Open access data from the Norwegian Statistical Institute containing updated demographic information were utilized in the calculation of point prevalence for each county in the southeast Norway health region. Ethical aspects
The Regional Committee of Medical Ethics in Southern Norway (REK sør), the Norwegian Ministry of Health (the Norwegian Patient Registry) and the Privacy Policy Department at OUH have all approved this study with all aspects related to patient data recording and ethical aspects related to the handling of patient sensitive material. Statistical analysis
Statistical analysis was undertaken with SPSS, version 20/21 (IBM, Armonk, NY, USA). For descriptive statistics, continuous variables with normal distribution are presented as mean with SD. Categorical variables are presented as numbers and percentages. Group differences were analysed by Student’s t-test (two-tailed,
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
unpaired). Pearson’s exact test and the chi-squared test were utilized for the comparison of independent groups of categorical data; the significance level was P < 0.05.
Results Study cohort
The combined case finding strategy identified 3160 patients with ICD-10 codes compatible with IIM (including sIBM), and a largely overlapping cohort of 500 patients with muscle biopsies encoded with inflammation (Fig. 1). A detailed review of the patient charts and muscle biopsy reports (see below) showed that 100 of the patients identified by the case finding strategies met either the 1997 and/or the 2011 ENMC criteria for sIBM. The 1997 criteria were met by 92 patients, whilst 95 met the 2011 ENMC criteria (Fig. 1). Altogether, 89 of the 100 patients meeting one or both sets of ENMC criteria were primarily identified by the ICD-10 search strategy, whilst 11 were captured by the neuropathology database search only (Fig. 1). These 11 patients had clinical sIBM features but had received ICD codes not included in our search: either M35.8 (specific connective tissue disease) at OUH or G72.9 (unspecified myopathy) at a local hospital. Overview of the items assessed by the ENMC criteria
All the 100 sIBM patients identified had sporadic disease of more than 1-year duration and a slowly progressive disease course. A retrospective chart review
Figure 1 Flow chart describing the search strategy applied for identification of the sporadic inclusion body myositis patients in southeast Norway, and the results of the broad search strategy.
© 2014 EAN
675
showed that all the patients had proximal muscle weakness and knee extension weakness. Distal weakness (i.e. finger flexor weakness) was described in 94/ 100 patients (Table 1). A detailed review of the primary muscle biopsy reports was performed in 97 patients (two patients only had chart information on histology, whilst one patient had end-stage muscle disease that was not suitable for detailed review). The most frequent biopsy features were endomysial inflammatory infiltrates (in 92/97 patients) and rimmed vacuoles (91/97). Upregulation of MHC 1 (of any degree) was described in 78/ 95 patients, with generalized MHC 1 upregulation in 61 biopsies. Electron microscopy was performed in 70 patients, and 16 of these had 15–18 nm tubulofilaments (Table 1). In addition, in 29 biopsies the possibility of muscle dystrophy was mentioned. In these cases, staining for common muscle dystrophy markers had been performed and described as normal (data not shown). Altogether, the neuropathology review showed that 79 patients had probable or definite sIBM histology features by the Griggs criteria [16]. The other 21 patients (all of whom had clinical sIBM) displayed histology features that were not inconsistent with sIBM (Fig. 1). Patient characteristics
At diagnosis, the mean age of sIBM patients was 66.9 years, and the mean time from symptom onset to diagnosis was 5.6 years (Table 2). The male to female ratio was 1.5:1. Clinical characteristics showed no gender bias, except for dysphagia which was more common in women (94%) than in men (65%) (Table 2).
676
G. C. DOBLOUG ET AL.
Table 1 Overview of the number of patients in the sporadic inclusion body myositis (sIBM) study cohort that scored positive for the individual items in the 1997 and 2011 ENMC criteria for sIBM ENMC criteria 1997 Clinical items, n/N 1 Proximal weakness 2 Distal weakness (i.e. finger flexion weakness)a 3a Finger flexion weakness > shoulder abduction weakness 3b Knee extension weakness ≥ hip flexion weakness 4 Slowly progressive course and sporadic disease 5 Duration >12 months 6 Age at onset >45 years 7 CK no greater than 15/ULN Histological items, n/N (%)b 8 Endomysial inflammatory infiltrates 9 Rimmed vacuoles 10 Protein accumulations or 15–18 mm filaments 11 Upregulation of MHC class 1 Scoring of ENMC criteria, n/N Definite sIBM (1, 2, 4, 8, 9 or 1, 4, 8, 9, 10) Probable sIBM (1, 2, 4, 8 or 1, 4, 8, 9) Clinico-pathologically defined sIBM (5 and 8–11) Clinically defined sIBM (3a and 3b, 5–7, any of 8–11) Probable sIBM (3a or 3b, 5–7, any of 8–11)
2011
100/100 94/100 n.a.a 100/100 100/100 100/100 95/100 100/100 92/97 (92) 91/97 (94) 16/70 (23) 78/95 (82) 84/92 8/92 16/95 n.a.a 79/95
CK, creatine kinase; ULN, upper limits of normal. CK ULN is 210 U/l in females, 400 U/l in males 0–49 years and 280 U/l in males >50 years. Finger flexion weakness was not quantified but registered as present or absent; bthe pathological analysis included the following stains: H&E, ATP-ase 4.3, ATP-ase 9.4, Gomori, NADH, succinat dehydrogenase, combined COX/succinat dehydrogenase stain, esterase, acid phosphatase, Oil Red O and PAS. The only immunohistochemical analysis performed was MHC 1 stain at Rikshospitalet and HLA stain at Ulleval. Immunohistochemical staining: simple Congo stain, p62 and TDP43 not available.
a
The mean maximum CK level recorded was 804 U/l and the mean ESR level was 28 mm. Serum autoantibodies were frequent, with ANA in 37% and antiSSA in 22%, with a female predominance (Table 1). Myositis specific antibodies (MSA) were positive in seven patients, of whom six were male. A review of comorbidity showed that heart disease, most commonly ischaemic heart disease, was present in 72% of the patients, with a male predominance (P = 0.007). Twelve female and 17 male patients died during the observation period, giving a total mortality rate of 29% (Table 2).
counties in the area were noted with the highest figure (70/1 000 000) observed in the county of Telemark (Fig. 2). The number of newly diagnosed sIBM cases in the study area proved to be relatively stable from year to year across the study period from 2003 to 2012 (Fig. 3a). Based on these retrospective data, estimated annual incidence rates of sIBM across the study period ranged from 2/1 000 000 to 6/1 000 000. The incidence appeared to peak in the age group 60–69 years with 42% of the patients being diagnosed in this age range (Fig. 3b).
Point prevalence and incidence of sIBM
Patient follow-up, treatment and supplementary analyses
By 31 December 2012, 71 of the 100 patients meeting at least one of the two sIBM criteria sets were still alive and living in the study area. Analysis by criteria set showed that 66 of the 92 patients meeting the 1997 ENMC criteria and 66 of the 95 patients meeting the 2011 criteria were alive. The estimated point prevalences of sIBM in southeast Norway by the 1997 and 2011 criteria were thus similar at 33/1 000 000. The point prevalence of patients meeting at least one of the two criteria sets was only slightly higher, at 35/ 1 000 000, but large differences between the individual
Follow-up of the sIBM patients was either by neurologists (n = 35) or rheumatologists (n = 65). Patients with coexisting rheumatic diseases (24%), most commonly Sjøgren’s syndrome (10%), were exclusively followed by rheumatologists (Table 3). Differences in treatment practice between specialists were seen: e.g. methotrexate was mostly prescribed by rheumatologists (Table 3). MRI and oesophagus imaging was more frequently utilized by rheumatologists than by neurologists (Table 3). MRI of thigh muscles had
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
677
Table 2 Clinical characteristics of the total sporadic inclusion body myositis (sIBM) cohort
Age at diagnosis, (years, mean, SD) Disease duration (years, mean, SD) a Diagnostic delay (years, mean, SD) Mortality, n/N (%) Myalgia, n/N (%) Arthralgia, n/N (%) Dysphagia, n/N (%) Sicca, n/N (%) CK max (U/ml, mean, SD) ESR (mm, mean, SD) ANA, n/N (%) Anti-SSA, n/N (%) MSA, n/N (%)
Total sIBM N = 100
Females N = 40
Males N = 60
P value
66.9 5.5 5.6 31/100 55/96 20/100 68/89 16/91 804 28 29/79 15/68 6/59
67.1 5.6 5.9 12/40 22/38 9/39 33/35 7/36 746 32 14/31 10/26 1/21
66.8 5.5 5.4 19/60 33/58 11/57 35/54 9/55 840 26 15/48 5/42 5/38
0.900 0.960 0.640 0.860 0.920 0.650 0.000 0.680 0.640 0.460 0.211 0.010 0.309
(9.3) (4.7) (5.0) (31) (57) (20) (76) (18) (777) (23) (37) (22) (10)
(9.5) (5.2) (5.2) (30) (58) (23) (94) (19) (907) (24) (45) (39) (5)
(9.3) (4.4) (4.8) (32) (57) (19) (65) (16) (695) (23) (31) (12) (13)
CK, creatine kinase, upper reference values 210 U/l in females and 280 U/l in males >50 years; ESR, erythrocyte sedimentation rate; ANA, anti-nuclear autoantibody; anti-SSA, anti-Sjøgren syndrome A autoantibody; MSA, myositis specific antibodies. a Diagnostic delay, measured in years: the time period from first onset of symptoms to diagnosis. Bold values represent significant P value (P < 0.05).
Figure 2 Total number of patients diagnosed with sporadic inclusion body myositis (sIBM) in each of the 10 counties of southeast Norway, and the estimated point prevalence of sIBM in each county by 31 December 2012.
been done in 56/100 patients, and 88% had atrophy and oedema. Oesophageal dysmotility was recorded in 45/54 patients examined by this modality. High resolution computed tomography of the lungs had been undertaken in 38 patients, with some degree of interstitial lung disease noted in 11 cases (Table 3). Five of these 11 patients were anti-SSA positive and had coexisting Sjøgren’s syndrome.
Discussion Overall, little is known about the occurrence of sIBM in the general population and unbiased data on the frequencies of key clinical and laboratory features are missing. Here, multiple acquisition routes were applied
© 2014 EAN
to identify all the sIBM cases in southeast Norway over a 10-year period. With this strategy, the point prevalence of sIBM in Norway at study end was estimated to be 33/1 000 000 by the 1997 and 2011 ENMC criteria. This estimate is substantially higher than previous European estimates [20,26,28]. There are important differences between the current study and previous sIBM epidemiology reports. First, since it was performed in an area with a denominator population of 2.6 million, it produced a large-sized, unselected sIBM cohort. Secondly, the sIBM work was done in conjunction with a study on the other IIM (PM and DM). This gave us the opportunity to search for sIBM, which does not have a specific ICD10 code, across a wide range of relevant myositis and
678
G. C. DOBLOUG ET AL.
myopathy codes; and it forced us to actively select against the possibility of PM and/or DM diagnoses in every potential sIBM case. The fact that the wide ICD-10 searches identified 89 of the 100 patients in the final sIBM cohort indicates that this approach was successful. Thirdly, the study had a long (10-year) acquisition period. Combined with manual review of chart data spanning from disease onset and onwards, this resulted in a long clinical observation period in the majority of the patients. Unfortunately, the number of sIBM patients that were initially diagnosed as PM or unspecific myositis was not formally recorded, but our impression from the chart review process was that approximately a fifth of the sIBM patients had another myositis diagnosis before they developed typical sIBM features and received this diagnosis. Fourthly, the study combined clinical data acquisition with an independent and complete review of muscle biopsy reports. The systematic review of the histology reports was important as it improved the ENMC histopathology scoring quality and allowed us to identify 11 additional sIBM cases that were not ICD-10 coded as myositis by the clinicians. The major limitation of the current study was that data acquisition was based on retrospective review of medical records, with missing and lacking documentation. The number of potential sIBM cases rejected due to missing chart review information was low, however. Patients were classified by two sIBM criteria sets, the ENMC criteria from 1997 and 2011 [17,19]. The study was primarily designed to capture items in the
(a)
(b)
Figure 3 (a) Age distribution of the patients, at the time of the diagnosis, in the sporadic inclusion body myositis (sIBM) study cohort. (b) Retrospective analysis on the number of new cases with sIBM diagnosed per year across the study period from 2003 to 2012.
Table 3 Overview of the frequencies of inflammatory rheumatic diseases, immune-modulating treatments and imaging analyses in the sporadic inclusion body myositis (sIBM) study cohort
Total (N = 100) Inflammatory rheumatic disease, n/N (%) Total Sjøgren’s syndrome Rheumatoid arthritis Systemic lupus erythematosus Treatment, n/N (%) Prednisolone Methotrexate Azathioprine Intravenous immunoglobulin Radiological studies, n/N (%) Muscle MRI performed Abnormal muscle MRI Dynamic X-ray oesophagus Abnormal X-ray oesophagus High resolution lung CT Abnormal lung high resolution CT
Followed at Department of Neurology (N = 35)
24/100 10/100 5/100 3/100
(24) (10) (5) (3)
2/35 (6) 0 0 0
73/97 23/96 11/95 32/96
(75) (24) (12) (33)
19/34 1/33 2/33 10/33
56/100 49/57 54/88 45/54 38/100 11/38
(56) (86) (61) (83) (38) (29)
10/35 7/10 9/35 7/9 14/35 2/14
Followed at Department of Rheumatology (N = 65)
P value
22/65 10/65 5/65 3/65
(34) (15) (8) (5)
0.002 0.014 0.092 0.200
(56) (3) (6) (30)
55/65 22/65 9/64 22/65
(85) (34) (14) (35)
0.002 0.001 0.240 0.724
(29) (70) (26) (78) (40) (14)
47/65(72) 42/47 (89) 45/53(85) 38/45 (84) 24/65(37) 9/24 (38)
0.000 0.110 0.000 0.624 0.762 0.128
MRI, magnetic resonance imaging; CT, computed tomography. Bold values represent significant P value (P < 0.05).
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
1997 criteria. These criteria were chosen because they were applied in the only previous, population-based sIBM study from Europe [20]. Shortly after data acquisition was completed, the new 2011 ENMC criteria were published. Since these criteria focus more on clinical sIBM features than the 1997 criteria, it was interesting to try to compare these criteria sets. Five patients who met the 1997 criteria were aged 30–45 at diagnosis and were therefore excluded by the 2011 criteria (Table 2). More interestingly, it was found that eight patients excluded by the 1997 criteria due to incomplete histology features met the 2011 criteria. Our data support the notion that diagnostic delay is common in sIBM. In fact, a delay (5.6 years) longer than recently reported from Oxford and Paris (4.9 years) [10] and Western Australia (4.4 years) [21] was observed, but it should be mentioned that in slowly progressive disease the age of onset is notoriously unreliable. Nonetheless, with new and possibly efficient treatments being tested [29], it will be critical to identify sIBM cases earlier. Surprisingly, rimmed vacuoles were described in 92 of the 97 primary muscle histology reports reviewed for the current study. It is not known why rimmed vacuoles were so frequent, but the fact that the reports were written by five different pathologists argues against reporter bias. The archetypical findings of inclusion body filaments were only found in a low number of our sIBM cases, but comparably low frequencies have been noted in many other clinical studies [19,30,31]. Interestingly, dysphagia was very frequently reported in our cohort, particularly in females, and oesophageal dysmotility was present in 83% of those who were examined by radiology. These findings are in line with many previous studies [11–13], but higher than in the large Oxford/Paris study where only 46% complained of dysphagia [10]. Available data suggest that the prevalence of sIBM varies considerably across populations and ethnic groups. In Caucasians, sIBM is associated with the HLA-DR3 gene and the corresponding 8.1 MHC ancestral haplotype, which is associated with several autoimmune diseases including primary Sjøgren’s syndrome [32–34]. It is speculated that the high frequencies of Sjøgren’s syndrome (10%) and anti-SSA (23%) in our cohort may reflect this common genetic association [32–34]. Interestingly, the prevalence of inflammatory rheumatic disorders (24%) in our cohort was twice as high as previously reported [34]. This high frequency may be explained, at least partly, by the fact that Norwegian sIBM patients are often primarily cared for by rheumatologists and that the HLADRB1*03 allele is common in Norway [35]. The rheumatology perspective may also explain the relatively
© 2014 EAN
679
high frequency of myalgia and arthralgia recorded, and the wide use of immuno-modulating medications compared with previous studies. Although treatment effects were not examined specifically, our impression was that the drugs applied did not lead to sustained improvement in muscle strength. The high prevalence of sIBM in this study led us to assess all aspects that may have led to an over-estimation of the prevalence. The contingency of familial or hereditary cases being included due to lack of knowledge might be a possibility, but this would not amount to large numbers. The relatively high frequency of the HLA-DRB1*03 allele in Norway (13.1% allele frequency [35], giving a total DR3 population frequency of about 25%) may possibly explain why sIBM is so frequent but, notably, there are no studies on the HLA genetics of sIBM in any of the Scandinavian countries. In conclusion, an sIBM point prevalence of 35/ 1 000 000 was observed. This prevalence is higher than previously noted in Europe, and our belief is that it is a close approximation of the true population prevalence of sIBM in Norway. Our results confirm that diagnostic delay is a major challenge in sIBM. This strongly suggests that measures to increase disease awareness are warranted.
Acknowledgements The authors thank Cecilie Kaufmann and Ase Lexberg, Department of Rheumatology, Buskerud Hospital, Vestre Viken, Olav Bjørneboe, Department of Rheumatology, Martina Hansens Hospital, Bærum, Patrick Stolt, Department of Rheumatology, Innlandet Hospital, Kongsvinger, Knut Mikkelsen, Department of Rheumatology, Innlandet Hospital, Lillehammer, Christian Gulseth, Department of Rheumatology, Betanien Hospital, Telemark, Anne Noraas Bendvold, Department of Rheumatology, Sørlandet Hospital, Kristiansand, Østfold, Tormod Fladby, Department of Neurology, Ahus Hospital, Akershus, Remo Gerdts, Department of Neurology, Vestfold Hospital, Vestfold, Sharka Øygaarden, Telemark Hospital, Telemark, Grethe Kleveland, Department of Neurology, Innlandet Hospital, Lillehammer and Hedmark, and Anne-Kathrine Palacios, Department of Neurology, Østfold Hospital, Østfold, for providing access to data. This work has been supported by grants from the Norwegian Women’s Public Health Association and Extrastiftelsen.
Disclosure of conflicts of interest The authors declare no financial or other conflicts of interest.
e41
G. C. DOBLOUG ET AL.
References 1. Yunis EJ, Samaha FJ. Inclusion body myositis. Lab Invest 1971; 25: 240–248. 2. Adams RD, Kakulas BA, Samaha FA. A myopathy with cellular inclusions. Trans Am Neurol Assoc 1965; 90: 213–216. 3. Chou SM. Myxovirus-like structures in a case of human chronic polymyositis. Science 1967; 158: 1453–1455. 4. Dalakas MC. Polymyositis, dermatomyositis and inclusion-body myositis. N Engl J Med 1991; 325: 1487–1498. 5. Amato AA, Barohn RJ. Inclusion body myositis: old and new concepts. J Neurol Neurosurg Psychiatry 2009; 80: 1186–1193. 6. Finch CE. A perspective on sporadic inclusion-body myositis: the role of aging and inflammatory processes. Neurology 2006; 66(2 Suppl 1): S1–S6. 7. Uruha A, Nishino I. [Pathogenesis of inclusion body myositis: autoimmune or degenerative disease?]. Brain Nerve 2013; 65: 1291–1298. 8. Roos PM, Vesterberg O, Nordberg M. Inclusion body myositis in Alzheimer’s disease. Acta Neurol Scand 2011; 124: 215–217. 9. Askanas V, Engel WK. Inclusion-body myositis: musclefiber molecular pathology and possible pathogenic significance of its similarity to Alzheimer’s and Parkinson’s disease brains. Acta Neuropathol 2008; 116: 583–595. 10. Benveniste O, Guiguet M, Freebody J, et al. Long-term observational study of sporadic inclusion body myositis. Brain 2011; 134(Pt 11): 3176–3184. 11. Cox FM, Titulaer MJ, Sont JK, et al. A 12-year followup in sporadic inclusion body myositis: an end stage with major disabilities. Brain 2011; 134(Pt 11): 3167– 3175. 12. Mulcahy KP, Langdon PC, Mastaglia F. Dysphagia in inflammatory myopathy: self-report, incidence, and prevalence. Dysphagia 2012; 27: 64–69. 13. Cox FM, Verschuuren JJ, Verbist BM, et al. Detecting dysphagia in inclusion body myositis. J Neurol 2009; 256: 2009–2013. 14. Hohlfeld R. Update on sporadic inclusion body myositis. Brain 2011; 134(Pt 11): 3141–3145. 15. Needham M, Mastaglia FL. Inclusion body myositis: current pathogenetic concepts and diagnostic and therapeutic approaches. Lancet Neurol 2007; 6: 620–631. 16. Griggs RC, Askanas V, DiMauro S, et al. Inclusion body myositis and myopathies. Ann Neurol 1995; 38: 705–713. 17. Verschuuren JJBU, Van Engelen BGM, et al. Inclusion body myositis. London: Royal Society of Medicine Press; 1997. 18. Benveniste O, Hilton-Jones D. International Workshop on Inclusion Body Myositis held at the Institute of Myology, Paris, on 29 May 2009. Neuromuscul Disord 2010; 20: 414–421. 19. Rose MR. 188th ENMC International Workshop: Inclusion Body Myositis, 2–4 December 2011, Naarden, The Netherlands. Neuromuscul Disord 2013; 23: 1044–1055. 20. Badrising UA, Maat-Schieman M, van Duinen SG, et al. Epidemiology of inclusion body myositis in the Netherlands: a nationwide study. Neurology 2000; 55: 1385–1387.
21. Phillips BA, Zilko PJ, Mastaglia FL. Prevalence of sporadic inclusion body myositis in Western Australia. Muscle Nerve 2000; 23: 970–972. 22. Needham M, Corbett A, Day T, et al. Prevalence of sporadic inclusion body myositis and factors contributing to delayed diagnosis. J Clin Neurosci 2008; 15: 1350– 1353. 23. Tan JA, Roberts-Thomson PJ, Blumbergs P, et al. Incidence and prevalence of idiopathic inflammatory myopathies in South Australia: a 30-year epidemiologic study of histology-proven cases. Int J Rheum Dis 2013; 16: 331–338. 24. Nakanishi H, Koike H, Matsuo K, et al. Demographic features of Japanese patients with sporadic inclusion body myositis: a single-center referral experience. Intern Med 2013; 52: 333–337. 25. Suzuki N, Aoki M, Mori-Yoshimura M, et al. Increase in number of sporadic inclusion body myositis (sIBM) in Japan. J Neurol 2012; 259: 554–556. 26. Oflazer PS, Deymeer F, Parman Y. Sporadic-inclusion body myositis (s-IBM) is not so prevalent in Istanbul/ Turkey: a muscle biopsy based survey. Acta Myol 2011; 30: 34–36. 27. Dobloug C, Garen T, Molberg O, et al. Prevalence and clinical characteristics of adult polymyositis and dermatomyositis; data from a large and unselected Norwegian cohort. Ann Rheum Dis 2014. doi: 10.1136/annrheumdis2013-205127. [Epub ahead of print]. 28. Weitoft T. Occurrence of polymyositis in the county of Gavleborg, Sweden. Scand J Rheumatol 1997; 26: 104– 106. 29. Lach-Trifilieff E, Minetti GC, Sheppard K, et al. An antibody blocking activin type II receptors induces strong skeletal muscle hypertrophy and protects from atrophy. Mol Cell Biol 2014; 34: 606–618. 30. Brady S, Squier W, Hilton-Jones D. Clinical assessment determines the diagnosis of inclusion body myositis independently of pathological features. J Neurol Neurosurg Psychiatry 2013; 84: 1240–1246. 31. Chahin N, Engel AG. Correlation of muscle biopsy, clinical course, and outcome in PM and sporadic IBM. Neurology 2008; 70: 418–424. 32. Mastaglia FL, Needham M, Scott A, et al. Sporadic inclusion body myositis: HLA-DRB1 allele interactions influence disease risk and clinical phenotype. Neuromuscul Disord 2009; 19: 763–765. 33. Needham M, James I, Corbett A, et al. Sporadic inclusion body myositis: phenotypic variability and influence of HLA-DR3 in a cohort of 57 Australian cases. J Neurol Neurosurg Psychiatry 2008; 79: 1056–1060. 34. Rojana-udomsart A, Needham M, Luo YB, et al. The association of sporadic inclusion body myositis and Sjogren’s syndrome in carriers of HLA-DR3 and the 8.1 MHC ancestral haplotype. Clin Neurol Neurosurg 2011; 113: 559–563. 35. Nordang GB, Flam ST, Maehlen MT, Kvien TK, Viken MK, Lie BA. HLA-C alleles confer risk for anti-citrullinated peptide antibody-positive rheumatoid arthritis independent of HLA-DRB1 alleles. Rheumatology (Oxford) 2013; 52: 1973–1982.
© 2014 EAN
Copyright of European Journal of Neurology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
High prevalence of inclusion body myositis in Norway; a population-based clinical epidemiology study €rnee, L. Grøvlef, J. T. Grana G. C. Doblouga, E. A. Antalb, L. Svebergc, T. Garena, H. Bitterd, J. Stja a,g and Ø. Molberg
EUROPEAN JOURNAL OF NEUROLOGY
a Department of Rheumatology, Oslo University Hospital (OUH), Oslo; bDepartment of Pathology, OUH, Oslo; cDepartment of Neurology, OUH, Oslo; dDepartment of Rheumatology, Sørlandet Hospital, Kristiansand; eDepartment of Rheumatology, Betanien Hospital, Skien; fDepartment of Rheumatology, Sykehuset Østfold, Moss; and gInstitute of Clinical Medicine, University of Oslo, Oslo, Norway
Keywords:
epidemiology, European Neuro-Muscular Centre Criteria, idiopathic inflammatory myopathy, sporadic inclusion body myositis Received 2 September 2014 Accepted 17 October 2014 European Journal of Neurology 2015, 22: 672–680 doi:10.1111/ene.12627
Background and purpose: Knowledge about the occurrence of sporadic inclusion body myositis (sIBM) in the general population is limited. Here, our aim was to identify and characterize every sIBM patient living in southeast Norway (population 2.64 million) from 2003 to 2012. Method: Two sIBM case finding strategies were applied. First, all hospital databases in southeast Norway were screened to identify cases with sIBM-compatible International Classification of Diseases 10 (ICD-10) codes. These cases were then manually chart reviewed. Secondly, all muscle histology reports encoded with inflammation were independently reviewed. Finally, cases were classified according to the 1997 and the 2011 European Neuro-Muscular Centre (ENMC) Research Diagnostic Criteria for sIBM. Results: The combined case finding strategy identified 3160 patients with sIBM compatible ICD-10 codes, and a largely overlapping cohort of 500 patients having muscle biopsies encoded with inflammation. Detailed retrospective review of chart and histology data showed that 95 patients met the 2011 ENMC sIBM criteria and 92 met the 1997 criteria. Estimated point prevalence of sIBM was 33/1 000 000, equal with both criteria sets. Mean age at diagnosis was 66.9 years and mean diagnostic delay was 5.6 years. Chart review revealed higher frequencies of dysphagia (94% vs. 65%) and anti-Sjøgren syndrome A antibodies (39% vs. 12%) in female sIBM patients (n = 40) than in males. Coexisting rheumatic diseases were present in 25% of sIBM cases, with Sjøgren’s syndrome in 10%. Conclusion: An estimated point prevalence of sIBM seven times higher than previously observed in Europe is reported. Our data show considerable diagnostic delay, a major challenge with new sIBM treatments in the pipeline.
Introduction Sporadic inclusion body myositis (sIBM) was first acknowledged as an entity separate from polymyositis (PM) in 1971 [1], but descriptions of histopathological and clinical features compatible with sIBM exist from the mid-1960s [2,3]. Traditionally, sIBM has been Correspondence: G. C. Dobloug, Department of Rheumatology, Oslo University Hospital, Postbox 4950, Nydalen, N-0424 Oslo, Norway (tel.: +47 23074878; fax: +47 22844651; e-mails: [email protected] or [email protected]).
672
regarded as one of the three idiopathic inflammatory myopathies (IIM), along with polymyositis (PM) and dermatomyositis (DM), and it may account for 30% of IIM cases [4,5]. There is growing evidence, however, that the disease mechanisms in sIBM differ from PM and DM [4] and represent a distinct separate entity with both inflammatory and degenerative changes [6–9]. Recent large-sized clinical studies provide further support to the notion that the muscle disease pattern in sIBM is highly characteristic, with asymmetric distribution, slow progression of proximal and distal
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
weakness and predominant involvement of the quadriceps and finger flexor muscles [10,11]. Dysphagia appears to be very frequent and can be disabling and potentially life-threatening [10,12,13]. Unlike PM and DM, sIBM is unresponsive to immune-modulating treatments. Hence, even though involvement of other organs than muscles is rare, sIBM has an end-stage of major disabilities [11,14]. There is no doubt that the light microscopy findings in sIBM muscle may be highly characteristic with coexisting processes suggestive of inflammation (endomysial infiltrates and lymphocyte invasion in morphologically normal myocytes) and degeneration (eosinophilic cytoplasmic inclusions and rimmed vacuoles) [15], but it is important to note that any given sIBM biopsy may lack any of these characteristic microscopic features [15]. Since the cause or pathogenesis of IBM is not known, there is no ‘gold standard’ for the diagnosis. There is an ongoing debate on how sIBM should optimally be diagnosed. The first widely used diagnostic criteria for sIBM, suggested by Griggs et al. in 1995 [16], were primarily based on histology findings. In 1997, the European Neuro-Muscular Centre (ENMC) produced a new criteria set which allowed for sIBM diagnosis in cases with highly suggestive clinical features but incomplete histology [17]. Diagnostic criteria primarily based on clinical findings were first proposed by Benveniste and Hilton-Jones in 2009 [18]. These criteria only required that histology was supportive and not inconsistent with sIBM [18]. The new ENMC diagnostic criteria (2011 ENMC IBM Research Diagnostic Criteria) [19] also emphasize clinical phenotype rather than pathology and additionally aim to allow for earlier diagnosis than previous criteria sets. Few studies on the epidemiology of sIBM have been performed. The first, large chart review study on sIBM was a nationwide collaborative cross-sectional study performed in the Netherlands in 1999 [20]. The study applied the 1997 ENMC criteria and identified 76 patients with sIBM, giving an estimated population prevalence of 4.9/1 000 000 [20]. The authors reported that prevalence was probably underestimated, but to date there are no equivalent European studies for comparison. Three Australian surveys, all based on local histopathology criteria, reported sIBM prevalence ranging from 9.3/1 000 000 to 50.5/1 000 000 [21–23]. In Japan, the prevalence of sIBM judged by the combination of several sIBM criteria sets [15,16,18] was estimated at 9.8/1 000 000 in 2003, with an increasing prevalence over the last decade [24,25]. In contrast, a recent biopsy-register study from Turkey reported only one sIBM case per million, possibly reflecting differences in sIBM prevalence across genetic backgrounds [26].
© 2014 EAN
673
Data on the occurrence of sIBM in the general population are limited. In the current study, our aim was to identify all sIBM cases in southeast Norway (with a denominator population of 2.64 million) and describe the clinical characteristics, laboratory findings and imaging data in the resulting cohort by retrospective chart and histology report review.
Materials and methods Study cohort
The sIBM study was part of a larger epidemiology study aiming to identify all IIM patients in southeast Norway. The PM/DM part of this IIM study was recently published [27]. Southeast Norway consists of 10 counties with 2 642 246 inhabitants and includes the largest cities in Norway. The main hospital in the region is Oslo University Hospital (OUH), which is the primary hospital for the city of Oslo (with 600 000 inhabitants) and the referral hospital for all the other hospitals in southeast Norway. In Norway, patients with IIM are followed by specialists at public hospitals, where sIBM patients are cared for either by neurologists or by rheumatologists. Patients included were residential and seen at hospitals in the health region between January 2003 and December 2012, a 10-year period. Study inclusion criteria
Patients were included if they fulfilled the following criteria: (i) disease classifiable as sIBM by the 1997 ENMC sIBM criteria and/or the 2011 ENMC IBM Research Diagnostic Criteria [17,19]; (ii) exclusion of PM or DM as possible diagnoses; (iii) inclusion body myositis not explained by familial disease; (iv) patient registered in the Norwegian Central Population Register with a home address in southeast Norway between 1 January 2003 and 31 December 2012. Case finding strategy
Two major acquisition routes were utilized to identify all the adult sIBM patients: (i) chart review of all potential IIM cases and (ii) retrospective review (performed by a neuropathologist) of all muscle biopsy reports encoded with inflammation. First, all potential IIM cases were identified by searches across patient administrative databases. Initially, the database at OUH was screened using the following International Classification of Diseases 10 (ICD-10) codes: M33.1 (dermatomyositis), M33.2 (polymyositis), M33.9 (unspecified PM/DM), M60.1
674
G. C. DOBLOUG ET AL.
(interstitial myositis), M60.8 (specified myositis), M60.9 (unspecified myositis), G72.4 (inflammatory myopathy, not classified elsewhere), G72.8 (other specified myopathies), G72.9 (unspecified myopathy), G73.7 (myopathy associated with disease classified elsewhere) [27]. Preliminary data from OUH showed that no sIBM cases were G-coded. Hence, the searches undertaken at the other southeast Norway hospitals were limited to the six M33 and M60 codes. The charts of all the patients identified were manually reviewed by the principal investigator (CD). The muscle biopsies taken in southeast Norway between 2003 and 2012 were examined by five neuropathologists at two laboratories in Oslo (Rikshospitalet and Ulleval) and one in Tromsø. From 2010, the two laboratories in Oslo were merged within the Department of Pathology, OUH. For our review, all the muscle histology reports encoded with inflammation in the Systematic Nomenclature of Medicine (SNOMED) code system were reviewed by a neuropathologist (EAA), and the following parameters were recorded; endomysial inflammatory infiltrates, lymphocyte invasion in structurally normal myocytes, rimmed vacuoles, MHC 1 expression (any and general) and the presence of inclusion body filaments at ultra-structural examination. The neuropathologist had access to muscle biopsy referral information and the pathology reports produced by the five other neuropathologists, but not the clinical charts. Finally, all the patients identified by chart review and/or muscle biopsy review were discussed on a case to case notion and scored, primarily according to the 1997 sIBM ENMC criteria and later by the 2011 ENMC IBM Research Diagnostic Criteria. Recording of patient data and items assessed by the 1997 and 2011 ENMC criteria
Predefined registration forms were used to record hospital chart data. Time of symptom onset and total disease duration, defined as the time from diagnosis to study end, or the time of death were recorded. Scoring of the clinical and histopathology items according to the 1997 and 2011 ENMC criteria was performed as follows: (i) proximal muscle weakness, described by a specialist (rheumatologist or neurologist) during clinical examination as weakness (and most often also atrophy) involving thigh and/or shoulder/neck muscles weakness was quantified by the Medical Research Council (MRC) scale (0–5) and/or manual muscle testing (MMT) performed by physiotherapists; (ii) distal muscle weakness, described by a specialist during clinical examination as weakness in the finger flexors and/or the combined presence of reduced grip strength
and atrophy of ulnar forearm muscles distal weakness was not quantified but was registered as present or absent; (iii) knee extension weakness, described by a specialist and quantified by MRC and/or MMT; (iv) slowly progressive course, chart description of weakness progressing slowly over years; (v) sporadic disease, no chart information on familial clustering; (vi) total disease duration over 12 months; (vii) age of onset above 30 years for 1997 criteria and above 45 for 2011 criteria; (viii) maximum creatine kinase (CK) levels, recorded from chart data; (ix) muscle biopsy parameters recorded from pathology reports: endomysial inflammatory infiltrates, lymphocyte invasion in structurally normal myocytes, rimmed vacuoles, MHC 1 expression (any and general), the presence of inclusion body filaments at ultra-structural examination. Other clinical information recorded from patient charts
Myalgia, arthritis (as defined by a doctor), arthralgia, dysphagia, sicca symptoms, Raynaud phenomenon, cough, dyspnoea, erythrocyte sedimentation rate (ESR), anti-nuclear antibodies (ANA), anti-Sjøgren syndrome A (SSA) antibodies, myositis specific antibodies (anti-Jo-1, PL-7, PL-12, SRP and Mi-2), electromyography/neurography, magnetic resonance imaging (MRI) of muscle, dynamic X-ray of the oesophagus and high resolution chest computed tomography were recorded. Occurrence of chronic heart and/or lung disease was also recorded. Time and cause of death was recorded when applicable. Open access data from the Norwegian Statistical Institute containing updated demographic information were utilized in the calculation of point prevalence for each county in the southeast Norway health region. Ethical aspects
The Regional Committee of Medical Ethics in Southern Norway (REK sør), the Norwegian Ministry of Health (the Norwegian Patient Registry) and the Privacy Policy Department at OUH have all approved this study with all aspects related to patient data recording and ethical aspects related to the handling of patient sensitive material. Statistical analysis
Statistical analysis was undertaken with SPSS, version 20/21 (IBM, Armonk, NY, USA). For descriptive statistics, continuous variables with normal distribution are presented as mean with SD. Categorical variables are presented as numbers and percentages. Group differences were analysed by Student’s t-test (two-tailed,
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
unpaired). Pearson’s exact test and the chi-squared test were utilized for the comparison of independent groups of categorical data; the significance level was P < 0.05.
Results Study cohort
The combined case finding strategy identified 3160 patients with ICD-10 codes compatible with IIM (including sIBM), and a largely overlapping cohort of 500 patients with muscle biopsies encoded with inflammation (Fig. 1). A detailed review of the patient charts and muscle biopsy reports (see below) showed that 100 of the patients identified by the case finding strategies met either the 1997 and/or the 2011 ENMC criteria for sIBM. The 1997 criteria were met by 92 patients, whilst 95 met the 2011 ENMC criteria (Fig. 1). Altogether, 89 of the 100 patients meeting one or both sets of ENMC criteria were primarily identified by the ICD-10 search strategy, whilst 11 were captured by the neuropathology database search only (Fig. 1). These 11 patients had clinical sIBM features but had received ICD codes not included in our search: either M35.8 (specific connective tissue disease) at OUH or G72.9 (unspecified myopathy) at a local hospital. Overview of the items assessed by the ENMC criteria
All the 100 sIBM patients identified had sporadic disease of more than 1-year duration and a slowly progressive disease course. A retrospective chart review
Figure 1 Flow chart describing the search strategy applied for identification of the sporadic inclusion body myositis patients in southeast Norway, and the results of the broad search strategy.
© 2014 EAN
675
showed that all the patients had proximal muscle weakness and knee extension weakness. Distal weakness (i.e. finger flexor weakness) was described in 94/ 100 patients (Table 1). A detailed review of the primary muscle biopsy reports was performed in 97 patients (two patients only had chart information on histology, whilst one patient had end-stage muscle disease that was not suitable for detailed review). The most frequent biopsy features were endomysial inflammatory infiltrates (in 92/97 patients) and rimmed vacuoles (91/97). Upregulation of MHC 1 (of any degree) was described in 78/ 95 patients, with generalized MHC 1 upregulation in 61 biopsies. Electron microscopy was performed in 70 patients, and 16 of these had 15–18 nm tubulofilaments (Table 1). In addition, in 29 biopsies the possibility of muscle dystrophy was mentioned. In these cases, staining for common muscle dystrophy markers had been performed and described as normal (data not shown). Altogether, the neuropathology review showed that 79 patients had probable or definite sIBM histology features by the Griggs criteria [16]. The other 21 patients (all of whom had clinical sIBM) displayed histology features that were not inconsistent with sIBM (Fig. 1). Patient characteristics
At diagnosis, the mean age of sIBM patients was 66.9 years, and the mean time from symptom onset to diagnosis was 5.6 years (Table 2). The male to female ratio was 1.5:1. Clinical characteristics showed no gender bias, except for dysphagia which was more common in women (94%) than in men (65%) (Table 2).
676
G. C. DOBLOUG ET AL.
Table 1 Overview of the number of patients in the sporadic inclusion body myositis (sIBM) study cohort that scored positive for the individual items in the 1997 and 2011 ENMC criteria for sIBM ENMC criteria 1997 Clinical items, n/N 1 Proximal weakness 2 Distal weakness (i.e. finger flexion weakness)a 3a Finger flexion weakness > shoulder abduction weakness 3b Knee extension weakness ≥ hip flexion weakness 4 Slowly progressive course and sporadic disease 5 Duration >12 months 6 Age at onset >45 years 7 CK no greater than 15/ULN Histological items, n/N (%)b 8 Endomysial inflammatory infiltrates 9 Rimmed vacuoles 10 Protein accumulations or 15–18 mm filaments 11 Upregulation of MHC class 1 Scoring of ENMC criteria, n/N Definite sIBM (1, 2, 4, 8, 9 or 1, 4, 8, 9, 10) Probable sIBM (1, 2, 4, 8 or 1, 4, 8, 9) Clinico-pathologically defined sIBM (5 and 8–11) Clinically defined sIBM (3a and 3b, 5–7, any of 8–11) Probable sIBM (3a or 3b, 5–7, any of 8–11)
2011
100/100 94/100 n.a.a 100/100 100/100 100/100 95/100 100/100 92/97 (92) 91/97 (94) 16/70 (23) 78/95 (82) 84/92 8/92 16/95 n.a.a 79/95
CK, creatine kinase; ULN, upper limits of normal. CK ULN is 210 U/l in females, 400 U/l in males 0–49 years and 280 U/l in males >50 years. Finger flexion weakness was not quantified but registered as present or absent; bthe pathological analysis included the following stains: H&E, ATP-ase 4.3, ATP-ase 9.4, Gomori, NADH, succinat dehydrogenase, combined COX/succinat dehydrogenase stain, esterase, acid phosphatase, Oil Red O and PAS. The only immunohistochemical analysis performed was MHC 1 stain at Rikshospitalet and HLA stain at Ulleval. Immunohistochemical staining: simple Congo stain, p62 and TDP43 not available.
a
The mean maximum CK level recorded was 804 U/l and the mean ESR level was 28 mm. Serum autoantibodies were frequent, with ANA in 37% and antiSSA in 22%, with a female predominance (Table 1). Myositis specific antibodies (MSA) were positive in seven patients, of whom six were male. A review of comorbidity showed that heart disease, most commonly ischaemic heart disease, was present in 72% of the patients, with a male predominance (P = 0.007). Twelve female and 17 male patients died during the observation period, giving a total mortality rate of 29% (Table 2).
counties in the area were noted with the highest figure (70/1 000 000) observed in the county of Telemark (Fig. 2). The number of newly diagnosed sIBM cases in the study area proved to be relatively stable from year to year across the study period from 2003 to 2012 (Fig. 3a). Based on these retrospective data, estimated annual incidence rates of sIBM across the study period ranged from 2/1 000 000 to 6/1 000 000. The incidence appeared to peak in the age group 60–69 years with 42% of the patients being diagnosed in this age range (Fig. 3b).
Point prevalence and incidence of sIBM
Patient follow-up, treatment and supplementary analyses
By 31 December 2012, 71 of the 100 patients meeting at least one of the two sIBM criteria sets were still alive and living in the study area. Analysis by criteria set showed that 66 of the 92 patients meeting the 1997 ENMC criteria and 66 of the 95 patients meeting the 2011 criteria were alive. The estimated point prevalences of sIBM in southeast Norway by the 1997 and 2011 criteria were thus similar at 33/1 000 000. The point prevalence of patients meeting at least one of the two criteria sets was only slightly higher, at 35/ 1 000 000, but large differences between the individual
Follow-up of the sIBM patients was either by neurologists (n = 35) or rheumatologists (n = 65). Patients with coexisting rheumatic diseases (24%), most commonly Sjøgren’s syndrome (10%), were exclusively followed by rheumatologists (Table 3). Differences in treatment practice between specialists were seen: e.g. methotrexate was mostly prescribed by rheumatologists (Table 3). MRI and oesophagus imaging was more frequently utilized by rheumatologists than by neurologists (Table 3). MRI of thigh muscles had
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
677
Table 2 Clinical characteristics of the total sporadic inclusion body myositis (sIBM) cohort
Age at diagnosis, (years, mean, SD) Disease duration (years, mean, SD) a Diagnostic delay (years, mean, SD) Mortality, n/N (%) Myalgia, n/N (%) Arthralgia, n/N (%) Dysphagia, n/N (%) Sicca, n/N (%) CK max (U/ml, mean, SD) ESR (mm, mean, SD) ANA, n/N (%) Anti-SSA, n/N (%) MSA, n/N (%)
Total sIBM N = 100
Females N = 40
Males N = 60
P value
66.9 5.5 5.6 31/100 55/96 20/100 68/89 16/91 804 28 29/79 15/68 6/59
67.1 5.6 5.9 12/40 22/38 9/39 33/35 7/36 746 32 14/31 10/26 1/21
66.8 5.5 5.4 19/60 33/58 11/57 35/54 9/55 840 26 15/48 5/42 5/38
0.900 0.960 0.640 0.860 0.920 0.650 0.000 0.680 0.640 0.460 0.211 0.010 0.309
(9.3) (4.7) (5.0) (31) (57) (20) (76) (18) (777) (23) (37) (22) (10)
(9.5) (5.2) (5.2) (30) (58) (23) (94) (19) (907) (24) (45) (39) (5)
(9.3) (4.4) (4.8) (32) (57) (19) (65) (16) (695) (23) (31) (12) (13)
CK, creatine kinase, upper reference values 210 U/l in females and 280 U/l in males >50 years; ESR, erythrocyte sedimentation rate; ANA, anti-nuclear autoantibody; anti-SSA, anti-Sjøgren syndrome A autoantibody; MSA, myositis specific antibodies. a Diagnostic delay, measured in years: the time period from first onset of symptoms to diagnosis. Bold values represent significant P value (P < 0.05).
Figure 2 Total number of patients diagnosed with sporadic inclusion body myositis (sIBM) in each of the 10 counties of southeast Norway, and the estimated point prevalence of sIBM in each county by 31 December 2012.
been done in 56/100 patients, and 88% had atrophy and oedema. Oesophageal dysmotility was recorded in 45/54 patients examined by this modality. High resolution computed tomography of the lungs had been undertaken in 38 patients, with some degree of interstitial lung disease noted in 11 cases (Table 3). Five of these 11 patients were anti-SSA positive and had coexisting Sjøgren’s syndrome.
Discussion Overall, little is known about the occurrence of sIBM in the general population and unbiased data on the frequencies of key clinical and laboratory features are missing. Here, multiple acquisition routes were applied
© 2014 EAN
to identify all the sIBM cases in southeast Norway over a 10-year period. With this strategy, the point prevalence of sIBM in Norway at study end was estimated to be 33/1 000 000 by the 1997 and 2011 ENMC criteria. This estimate is substantially higher than previous European estimates [20,26,28]. There are important differences between the current study and previous sIBM epidemiology reports. First, since it was performed in an area with a denominator population of 2.6 million, it produced a large-sized, unselected sIBM cohort. Secondly, the sIBM work was done in conjunction with a study on the other IIM (PM and DM). This gave us the opportunity to search for sIBM, which does not have a specific ICD10 code, across a wide range of relevant myositis and
678
G. C. DOBLOUG ET AL.
myopathy codes; and it forced us to actively select against the possibility of PM and/or DM diagnoses in every potential sIBM case. The fact that the wide ICD-10 searches identified 89 of the 100 patients in the final sIBM cohort indicates that this approach was successful. Thirdly, the study had a long (10-year) acquisition period. Combined with manual review of chart data spanning from disease onset and onwards, this resulted in a long clinical observation period in the majority of the patients. Unfortunately, the number of sIBM patients that were initially diagnosed as PM or unspecific myositis was not formally recorded, but our impression from the chart review process was that approximately a fifth of the sIBM patients had another myositis diagnosis before they developed typical sIBM features and received this diagnosis. Fourthly, the study combined clinical data acquisition with an independent and complete review of muscle biopsy reports. The systematic review of the histology reports was important as it improved the ENMC histopathology scoring quality and allowed us to identify 11 additional sIBM cases that were not ICD-10 coded as myositis by the clinicians. The major limitation of the current study was that data acquisition was based on retrospective review of medical records, with missing and lacking documentation. The number of potential sIBM cases rejected due to missing chart review information was low, however. Patients were classified by two sIBM criteria sets, the ENMC criteria from 1997 and 2011 [17,19]. The study was primarily designed to capture items in the
(a)
(b)
Figure 3 (a) Age distribution of the patients, at the time of the diagnosis, in the sporadic inclusion body myositis (sIBM) study cohort. (b) Retrospective analysis on the number of new cases with sIBM diagnosed per year across the study period from 2003 to 2012.
Table 3 Overview of the frequencies of inflammatory rheumatic diseases, immune-modulating treatments and imaging analyses in the sporadic inclusion body myositis (sIBM) study cohort
Total (N = 100) Inflammatory rheumatic disease, n/N (%) Total Sjøgren’s syndrome Rheumatoid arthritis Systemic lupus erythematosus Treatment, n/N (%) Prednisolone Methotrexate Azathioprine Intravenous immunoglobulin Radiological studies, n/N (%) Muscle MRI performed Abnormal muscle MRI Dynamic X-ray oesophagus Abnormal X-ray oesophagus High resolution lung CT Abnormal lung high resolution CT
Followed at Department of Neurology (N = 35)
24/100 10/100 5/100 3/100
(24) (10) (5) (3)
2/35 (6) 0 0 0
73/97 23/96 11/95 32/96
(75) (24) (12) (33)
19/34 1/33 2/33 10/33
56/100 49/57 54/88 45/54 38/100 11/38
(56) (86) (61) (83) (38) (29)
10/35 7/10 9/35 7/9 14/35 2/14
Followed at Department of Rheumatology (N = 65)
P value
22/65 10/65 5/65 3/65
(34) (15) (8) (5)
0.002 0.014 0.092 0.200
(56) (3) (6) (30)
55/65 22/65 9/64 22/65
(85) (34) (14) (35)
0.002 0.001 0.240 0.724
(29) (70) (26) (78) (40) (14)
47/65(72) 42/47 (89) 45/53(85) 38/45 (84) 24/65(37) 9/24 (38)
0.000 0.110 0.000 0.624 0.762 0.128
MRI, magnetic resonance imaging; CT, computed tomography. Bold values represent significant P value (P < 0.05).
© 2014 EAN
EPIDEMIOLOGY OF sIBM IN NORWAY
1997 criteria. These criteria were chosen because they were applied in the only previous, population-based sIBM study from Europe [20]. Shortly after data acquisition was completed, the new 2011 ENMC criteria were published. Since these criteria focus more on clinical sIBM features than the 1997 criteria, it was interesting to try to compare these criteria sets. Five patients who met the 1997 criteria were aged 30–45 at diagnosis and were therefore excluded by the 2011 criteria (Table 2). More interestingly, it was found that eight patients excluded by the 1997 criteria due to incomplete histology features met the 2011 criteria. Our data support the notion that diagnostic delay is common in sIBM. In fact, a delay (5.6 years) longer than recently reported from Oxford and Paris (4.9 years) [10] and Western Australia (4.4 years) [21] was observed, but it should be mentioned that in slowly progressive disease the age of onset is notoriously unreliable. Nonetheless, with new and possibly efficient treatments being tested [29], it will be critical to identify sIBM cases earlier. Surprisingly, rimmed vacuoles were described in 92 of the 97 primary muscle histology reports reviewed for the current study. It is not known why rimmed vacuoles were so frequent, but the fact that the reports were written by five different pathologists argues against reporter bias. The archetypical findings of inclusion body filaments were only found in a low number of our sIBM cases, but comparably low frequencies have been noted in many other clinical studies [19,30,31]. Interestingly, dysphagia was very frequently reported in our cohort, particularly in females, and oesophageal dysmotility was present in 83% of those who were examined by radiology. These findings are in line with many previous studies [11–13], but higher than in the large Oxford/Paris study where only 46% complained of dysphagia [10]. Available data suggest that the prevalence of sIBM varies considerably across populations and ethnic groups. In Caucasians, sIBM is associated with the HLA-DR3 gene and the corresponding 8.1 MHC ancestral haplotype, which is associated with several autoimmune diseases including primary Sjøgren’s syndrome [32–34]. It is speculated that the high frequencies of Sjøgren’s syndrome (10%) and anti-SSA (23%) in our cohort may reflect this common genetic association [32–34]. Interestingly, the prevalence of inflammatory rheumatic disorders (24%) in our cohort was twice as high as previously reported [34]. This high frequency may be explained, at least partly, by the fact that Norwegian sIBM patients are often primarily cared for by rheumatologists and that the HLADRB1*03 allele is common in Norway [35]. The rheumatology perspective may also explain the relatively
© 2014 EAN
679
high frequency of myalgia and arthralgia recorded, and the wide use of immuno-modulating medications compared with previous studies. Although treatment effects were not examined specifically, our impression was that the drugs applied did not lead to sustained improvement in muscle strength. The high prevalence of sIBM in this study led us to assess all aspects that may have led to an over-estimation of the prevalence. The contingency of familial or hereditary cases being included due to lack of knowledge might be a possibility, but this would not amount to large numbers. The relatively high frequency of the HLA-DRB1*03 allele in Norway (13.1% allele frequency [35], giving a total DR3 population frequency of about 25%) may possibly explain why sIBM is so frequent but, notably, there are no studies on the HLA genetics of sIBM in any of the Scandinavian countries. In conclusion, an sIBM point prevalence of 35/ 1 000 000 was observed. This prevalence is higher than previously noted in Europe, and our belief is that it is a close approximation of the true population prevalence of sIBM in Norway. Our results confirm that diagnostic delay is a major challenge in sIBM. This strongly suggests that measures to increase disease awareness are warranted.
Acknowledgements The authors thank Cecilie Kaufmann and Ase Lexberg, Department of Rheumatology, Buskerud Hospital, Vestre Viken, Olav Bjørneboe, Department of Rheumatology, Martina Hansens Hospital, Bærum, Patrick Stolt, Department of Rheumatology, Innlandet Hospital, Kongsvinger, Knut Mikkelsen, Department of Rheumatology, Innlandet Hospital, Lillehammer, Christian Gulseth, Department of Rheumatology, Betanien Hospital, Telemark, Anne Noraas Bendvold, Department of Rheumatology, Sørlandet Hospital, Kristiansand, Østfold, Tormod Fladby, Department of Neurology, Ahus Hospital, Akershus, Remo Gerdts, Department of Neurology, Vestfold Hospital, Vestfold, Sharka Øygaarden, Telemark Hospital, Telemark, Grethe Kleveland, Department of Neurology, Innlandet Hospital, Lillehammer and Hedmark, and Anne-Kathrine Palacios, Department of Neurology, Østfold Hospital, Østfold, for providing access to data. This work has been supported by grants from the Norwegian Women’s Public Health Association and Extrastiftelsen.
Disclosure of conflicts of interest The authors declare no financial or other conflicts of interest.
e41
G. C. DOBLOUG ET AL.
References 1. Yunis EJ, Samaha FJ. Inclusion body myositis. Lab Invest 1971; 25: 240–248. 2. Adams RD, Kakulas BA, Samaha FA. A myopathy with cellular inclusions. Trans Am Neurol Assoc 1965; 90: 213–216. 3. Chou SM. Myxovirus-like structures in a case of human chronic polymyositis. Science 1967; 158: 1453–1455. 4. Dalakas MC. Polymyositis, dermatomyositis and inclusion-body myositis. N Engl J Med 1991; 325: 1487–1498. 5. Amato AA, Barohn RJ. Inclusion body myositis: old and new concepts. J Neurol Neurosurg Psychiatry 2009; 80: 1186–1193. 6. Finch CE. A perspective on sporadic inclusion-body myositis: the role of aging and inflammatory processes. Neurology 2006; 66(2 Suppl 1): S1–S6. 7. Uruha A, Nishino I. [Pathogenesis of inclusion body myositis: autoimmune or degenerative disease?]. Brain Nerve 2013; 65: 1291–1298. 8. Roos PM, Vesterberg O, Nordberg M. Inclusion body myositis in Alzheimer’s disease. Acta Neurol Scand 2011; 124: 215–217. 9. Askanas V, Engel WK. Inclusion-body myositis: musclefiber molecular pathology and possible pathogenic significance of its similarity to Alzheimer’s and Parkinson’s disease brains. Acta Neuropathol 2008; 116: 583–595. 10. Benveniste O, Guiguet M, Freebody J, et al. Long-term observational study of sporadic inclusion body myositis. Brain 2011; 134(Pt 11): 3176–3184. 11. Cox FM, Titulaer MJ, Sont JK, et al. A 12-year followup in sporadic inclusion body myositis: an end stage with major disabilities. Brain 2011; 134(Pt 11): 3167– 3175. 12. Mulcahy KP, Langdon PC, Mastaglia F. Dysphagia in inflammatory myopathy: self-report, incidence, and prevalence. Dysphagia 2012; 27: 64–69. 13. Cox FM, Verschuuren JJ, Verbist BM, et al. Detecting dysphagia in inclusion body myositis. J Neurol 2009; 256: 2009–2013. 14. Hohlfeld R. Update on sporadic inclusion body myositis. Brain 2011; 134(Pt 11): 3141–3145. 15. Needham M, Mastaglia FL. Inclusion body myositis: current pathogenetic concepts and diagnostic and therapeutic approaches. Lancet Neurol 2007; 6: 620–631. 16. Griggs RC, Askanas V, DiMauro S, et al. Inclusion body myositis and myopathies. Ann Neurol 1995; 38: 705–713. 17. Verschuuren JJBU, Van Engelen BGM, et al. Inclusion body myositis. London: Royal Society of Medicine Press; 1997. 18. Benveniste O, Hilton-Jones D. International Workshop on Inclusion Body Myositis held at the Institute of Myology, Paris, on 29 May 2009. Neuromuscul Disord 2010; 20: 414–421. 19. Rose MR. 188th ENMC International Workshop: Inclusion Body Myositis, 2–4 December 2011, Naarden, The Netherlands. Neuromuscul Disord 2013; 23: 1044–1055. 20. Badrising UA, Maat-Schieman M, van Duinen SG, et al. Epidemiology of inclusion body myositis in the Netherlands: a nationwide study. Neurology 2000; 55: 1385–1387.
21. Phillips BA, Zilko PJ, Mastaglia FL. Prevalence of sporadic inclusion body myositis in Western Australia. Muscle Nerve 2000; 23: 970–972. 22. Needham M, Corbett A, Day T, et al. Prevalence of sporadic inclusion body myositis and factors contributing to delayed diagnosis. J Clin Neurosci 2008; 15: 1350– 1353. 23. Tan JA, Roberts-Thomson PJ, Blumbergs P, et al. Incidence and prevalence of idiopathic inflammatory myopathies in South Australia: a 30-year epidemiologic study of histology-proven cases. Int J Rheum Dis 2013; 16: 331–338. 24. Nakanishi H, Koike H, Matsuo K, et al. Demographic features of Japanese patients with sporadic inclusion body myositis: a single-center referral experience. Intern Med 2013; 52: 333–337. 25. Suzuki N, Aoki M, Mori-Yoshimura M, et al. Increase in number of sporadic inclusion body myositis (sIBM) in Japan. J Neurol 2012; 259: 554–556. 26. Oflazer PS, Deymeer F, Parman Y. Sporadic-inclusion body myositis (s-IBM) is not so prevalent in Istanbul/ Turkey: a muscle biopsy based survey. Acta Myol 2011; 30: 34–36. 27. Dobloug C, Garen T, Molberg O, et al. Prevalence and clinical characteristics of adult polymyositis and dermatomyositis; data from a large and unselected Norwegian cohort. Ann Rheum Dis 2014. doi: 10.1136/annrheumdis2013-205127. [Epub ahead of print]. 28. Weitoft T. Occurrence of polymyositis in the county of Gavleborg, Sweden. Scand J Rheumatol 1997; 26: 104– 106. 29. Lach-Trifilieff E, Minetti GC, Sheppard K, et al. An antibody blocking activin type II receptors induces strong skeletal muscle hypertrophy and protects from atrophy. Mol Cell Biol 2014; 34: 606–618. 30. Brady S, Squier W, Hilton-Jones D. Clinical assessment determines the diagnosis of inclusion body myositis independently of pathological features. J Neurol Neurosurg Psychiatry 2013; 84: 1240–1246. 31. Chahin N, Engel AG. Correlation of muscle biopsy, clinical course, and outcome in PM and sporadic IBM. Neurology 2008; 70: 418–424. 32. Mastaglia FL, Needham M, Scott A, et al. Sporadic inclusion body myositis: HLA-DRB1 allele interactions influence disease risk and clinical phenotype. Neuromuscul Disord 2009; 19: 763–765. 33. Needham M, James I, Corbett A, et al. Sporadic inclusion body myositis: phenotypic variability and influence of HLA-DR3 in a cohort of 57 Australian cases. J Neurol Neurosurg Psychiatry 2008; 79: 1056–1060. 34. Rojana-udomsart A, Needham M, Luo YB, et al. The association of sporadic inclusion body myositis and Sjogren’s syndrome in carriers of HLA-DR3 and the 8.1 MHC ancestral haplotype. Clin Neurol Neurosurg 2011; 113: 559–563. 35. Nordang GB, Flam ST, Maehlen MT, Kvien TK, Viken MK, Lie BA. HLA-C alleles confer risk for anti-citrullinated peptide antibody-positive rheumatoid arthritis independent of HLA-DRB1 alleles. Rheumatology (Oxford) 2013; 52: 1973–1982.
© 2014 EAN
Copyright of European Journal of Neurology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
