- Email: [email protected]
Macrophagic myofasciitis: an emerging entity
ARTICLES
Macrophagic myofasciitis: an emerging entity R K Gherardi, M Coquet, P Chérin, F-J Authier, P Laforêt, L Bélec, D Figarella-Branger, J-M Mussini, J-F Pellissier, M Fardeau, for the Groupe d’Études et Recherche sur les Maladies Musculaires Acquises et Dysimmunitaires (GERMMAD) de l’Association Française contre les Myopathies (AFM)
Summary Background An unusual inflammatory myopathy characterised by an infiltration of non-epithelioid histiocytic cells has been recorded with increasing frequency in the past 5 years in France. We reassessed some of these cases. Methods We did a retrospective analysis of 18 such cases seen in five myopathology centres between May, 1993, and December, 1997. The myopathological changes were reassessed at a clinopathology seminar. Findings Detailed clinical information was available for 14 patients. The main presumptive diagnoses were polymyositis and polymyalgia rheumatica. Symptoms included myalgias in 12 patients, arthralgias in nine, muscle weakness in six, pronounced asthenia in five, and fever in four. Abnormal laboratory findings were occasionally observed, and included raised creatine kinase concentrations, increased erythrocyte sedimentation rate, and myopathic electromyography. Muscle biopsy showed infiltration of the subcutaneous tissue, epimysium, perimysium, and perifascicular endomysium by sheets of large macrophages, with a finely granular PAS-positive content. Also present were occasional CD8 T cells, and inconspicuous muscle-fibre damage. Epithelioid and giant cells, necrosis, and mitotic figures were not seen. The images were easily distinguishable from sarcoid myopathy and fasciitis-panniculitis syndromes. Whipple’s disease, Mycobacterium avium intracellulare infection, and malakoplakia could not be confirmed. Ten patients were treated with various combinations of steroids and antibiotics; symptoms improved in eight patients, and stabilised in two. Interpretation A new inflammatory muscle disorder of unknown cause, characterised by a distinctive pathological pattern of macrophagic myofasciitis, is emerging in France.
Lancet 1998; 352: 347–52 Groupe d’Etude et de Recherche sur le Muscle et le Nerf (GERMEN), EA 2347, Université Paris XII-Val de Marne, Département de Pathologie, Hôpital Henri Mondor, F-94010 Créteil, France (Prof R K Gherardi MD, F-J Authier MD); Unité de Myopathologie, Départment d’Anatomie Pathologique, Centre Hospitalier Universitaire de Bordeaux, Hôpital Pellegrin, Bordeaux (M Coquet MD); Service de Médecine Interne et Institut de Myologie-INSERM U153 Groupe Hospitalier Pitié-Salpêtrière, Paris (P Cherin MD, P Laforêt MD, Prof M Fardeau MD); Service de Microbiologie, Hôpital Broussais, Paris (L Belec MD); Laboratoire de Biopathologie Nerveuse et Musculaire (JE 2053, Université Aix-Marseille II), Faculté de Médecine, Marseille (Prof D Figarella-Branger MD, Prof J-F Pellissier MD); and Laboratoire d’Anatomie Pathologique A, Bâtiment Jean Monnet, CHR 44000 Nantes (J-M Mussini MD) Correspondence to: Prof R K Gherardi (e-mail: [email protected])
THE LANCET • Vol 352 • August 1, 1998
Introduction Emerging diseases are challenging clinicians and pathologists with increasing frequency.1 Careful analysis of muscle biopsy samples have provided valuable evidence of the pathogenesis of idiopathic inflammatory myopathies,2 and myopathological alterations allow classification of these disorders into three main categories: dermatomyositis, polymyositis, and inclusionbody myositis. Other inflammatory myopathies3 include sarcoid and sarcoid-like granulomatous myopathies, and various autoimmune, infectious, or toxic diseases. The Spanish toxic-oil syndrome,4 for example, and the adulterated L-tryptophan-associated eosinophiliamyalgia syndrome5 are closely-related disorders6 associated with fasciitis, myositis, and systemic involvement, that appeared in 1981 and 1989, respectively. Other emerging inflammatory muscle diseases are those associated with HIV-1 infection, the incidence of which rose sharply after 1987,7 because of zidovudine myopathy, a toxic mitochondrial myopathy associated with myositis.8,9 We report the emergence of a new type of inflammatory myopathy of unkown origin, characterised by a unique myopathological pattern of infiltration of the epimysium, perimysium, and perifascicular endomysium by non-epithelioid periodic-acid-schiff (PAS)-positive cells of macrophage lineage. The disorder is easily distinguishable from sarcoid myopathy and fasciitis-panniculitis syndromes, was first seen in 1993, and has increased in frequency since that time in five French myopathological centres.
Methods From May, 1993, to December, 1997, an unusual infiltration of skeletal-muscle connective-tissue structures by non-epithelioid histiocytic cells was seen with increasing frequency in the myopathology centres of Bordeaux (three cases), Marseilles (three cases), Nantes (one case), and Paris (University Hospital Pitié-Salpêtriere, four cases, University Hospital Henri Mondor, Créteil, seven cases; figure 1). The yearly number of muscle biopsies examined in these myopathology centres is about 1200. All samples were collected as part of the routine tests in these laboratories. The patients were adults who had been referred with the presumptive diagnosis of polymyositis (ten cases), polymyalgia rheumatica (four cases), muscle dystrophy (one
Figure 1: Cases of macrophagic myofasciitis
347
ARTICLES
Case number
Sex/age (yr)
Duration of symptoms (mo)
1
F/55
6
2 3 4 5
M/38 M/48 F/70 F/56
11 36 3 24
6 7 8 9 10 11 12 13
M/35 F/37 M/61 M/53 F/29 M/25 F/30 F/52
4 14 12 48 24 12 6 7
14
F/29
3
Musculoskeletal symptoms
Other symptoms
EMG findings
CK
Myalgias, arthralgias, weakness, tenderness Myalgias, arthralgias, weakness Myalgias, arthralgias Myalgias, arthralgias Myalgias, arthralgias, weakness, tenderness Myalgias, arthralgias Myalgias, arthralgias Myalgias, arthralgias Weakness Myalgias Myalgias Myalgias, arthralgias, tenderness Myalgias, arthralgias, weakness
Ankle oedema
Myopathic
Normal
55
Remission (on treatment*)
Chest pain Ankle oedema None Headaches, fever, fatigue Fatigue None Cough, fever Dyspnoea, fatigue None Dyspnoea, fever, fatigue Dyspnoea, fever, fatigue Cough, peripheral neuropathy None
Not done Normal Normal Myopathic
⫻4 ⫻3 Normal ⫻100
12 10 110 115
Remission (on treatment†) Spontaneous remission Remission (on treatment†) Remission (on treatment†)
Normal Normal Not done Normal Myopathic Normal Myopathic Neuropathic
⫻1·5 Normal Normal Normal ⫻5 ⫻3 Normal Normal
Not done 10 70 Not done 4 8 26 6·5
Stabilisation (on treatment*) Remission (on treatment†) Remission (on treatment‡) Slow progression Remission (on treatment†) Slow progression Remission (on treatment†) Stabilisation (on treatment†)
Neuropathic
Normal
Not done
Slow progression
Weakness
ESR (mm/h)
Outcome
EMG=electromyogram; CK=creatine kinase; ESR=erythrocyte sedimentation rate. *Antibiotics combined with steroids. †Steroids alone. ‡Antibiotics alone.
Table 1: Clinical and biological findings in patients with macrophagic myofasciitis case), or mitochondrial cytopathy (three cases). Detailed clinical information was available for 14 of the 18 patients. The deltoid-muscle biopsy samples had been conventionally divided into three portions, frozen, and embedded in paraffin and epoxy, and had been processed in the standard way. Myopathological alterations were systematically reassessed by five myopathologists (RG, MC, J-MM, J-FP, MF) in September, 1997, during a clinicopathological seminar of the Groupe d’Etudes et Recherche sur les Maladies Musculaires Acquises et Dysimmunitaires (GERMMAD), a branch of the Association Française contre les Myopathies. Conclusions of the seminar were transmitted to the Réseau National de Santé Publique France, and to the Centers for Disease Control, USA.
Results Characteristics of patients are listed in table 1. 12 patients were white, one was West Indian, and one was Algerian. Four patients had stayed for long times in tropical countries (three in Africa, one in the West Indies), and four were working in a hospital. Four patients were taking no medication at the time of first muscular symptoms, and four were on various drugs not known to cause myopathy, four were on chloroquine to prevent malaria, one took hydroxychloroquine as an immunomodulator, and one was on lipid-lowering drugs that were withdrawn, but symptoms did not improve. Two patients had had tuberculosis and one had been cured of sarcoidosis. Medical history (from 1 year before first muscle symptoms) included a connective-tissue disease in three patients (systemic lupus erythematosus, rheumatoid arthritis, and Hashimoto’s thyroiditis), myelodysplasia in one, and uvula carcinoma treated by irradiation at time of first muscular symptoms in one. Clinical features and abnormal laboratory findings are summarised in tables 2 and 3. All patients had had Symptom
Number of patients (n=14)
Myalgias Arthralgias Muscle weakness Fatigue Fever Muscle tenderness Dyspnoea Acroparaesthesiae in hands Ankle oedema Dry cough Epigastric pain Chest pain Dysphagia Headache
12 9 6 5 4 4 3 2 2 2 2 1 1 1
Table 2: Symptoms of patients with macrophagic myofasciitis
348
muscular symptoms for 3–48 months before biopsy. Mild to moderate chronic myalgias were the most common manifestation, occurring in 12 (86%) of 14 patients. In two patients, muscle pain was localised in upper limbs (girdle, arms, and forearms), in four patients it was localised in upper limbs and thighs, and in six patients myalgias were diffuse. In three patients, myalgias started or increased on exertion. Arthralgias were present in nine (64%) of 14 patients, with large joints involved in all nine patients (shoulders, elbows, wrists, lumbosacral spine, or hips) and finger joints in three. Six of 14 patients (43%) complained of mild to moderate muscle weakness, with proximal muscles of all arms and legs affected in four patients, and selective involvement of lower limbs in two. Six (43%) of 14 patients had general symptoms, including marked asthenia (five) and fever ranging from 38ºC to 39ºC (four). Digestive symptoms other than epigastric pain were absent at presentation. One patient developed diarrhoea when on long-term antibiotic therapy. Physical examination showed no muscle wasting. Muscle tenderness was present in four (29%) of 14 patients, and muscle induration in two (14%). Skin changes, including peau d’orange induration, were not seen. Except for nocturnal paresthaesias in hands and a bilateral Tinel’s sign noted in two (14%) of 14 patients, there was no peripheral neuropathy. There was neither lymphadenopathy nor hepatosplenomegaly. Electromyography was myopathic in four (33%) of 12 patients and equivocal in two (17%) of 12. Study of nerve-conduction velocity showed evidence of sensoryperipheral-nerve involvement in two patients. Serum Test
Number of patients positive
Number of patients tested
Creatine kinase >200 IU/L Aldolase >8 IU/L AST/ALT >35 IU/L Erythrocyte sedimentation rate >40 mm/h C-reactive protein >0·8 mg/dL Haemoglobin <12 g/L Lymphocytes 1·0⫻109/L Monocytes 0·5⫻109/L Myopathic electromyography Abnormal muscle magnetic-resonance imaging
6 3 2 5 4 3 2 1 4 1
14 7 11 11 10 10 10 10 12 4
AST/ALT=aspartate aminotransferase/alanine aminotransferase. Autoantibodies detected in patient with known systemic lupus erythematosus, and rheumatoid factor detected in patient with rheumatoid arthritis were excluded.
Table 3: Abnormal laboratory findings in patients with macrophagic myofasciitis
THE LANCET • Vol 352 • August 1, 1998
ARTICLES
Case number
1 2 3 4 5 6 7 9 10 11 12 13 14
Macrophagic infiltrate
T-cell infiltrate
Epimysium
Perimysium
Endomysium
Giant cells
PAS-positive
Sparse
Perivascular
Myonecrosis
+++ +++ +++ ⫺ +++ +++ ++ ⫺ ++ ++ +++ +++ +++
+++ ++ ⫺ + +++ +++ ++ ++ +++ ++ ++ + ++
+++ + ⫺ +++ + ++ + + ++ + + ⫺ ++
⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
+ + ⫺ + ⫺ + + ⫺ + + ⫺ + +
⫺ + + + ++ ⫺ + ⫺ +++ + + ⫺ +
+/⫺ ⫺ +/⫺ ⫺ ⫺ ⫺ +/⫺ ⫺ ⫺ ⫺ ⫺ +/⫺ ⫺
⫺=absent; +/⫺=almost completely absent; +=mild; ++=moderate; +++=severe.
Table 4: Myopathological findings in patients with macrophagic myofasciitis
creatine-kinase concentration was raised in six (43%) of 14 patients, with increase of 1·5–5·0 times the normal in five patients, and more than a ten-fold increase in one. Aldolase concentration was slightly raised in three (43%) of seven patients, including one who had a normal creatine kinase concentration. Aspartateaminotransferase and alanine-aminotransferase were increased consistently with the raised creatine kinase concentrations in two (18%) of 11 patients. ␥-glutamyl transpeptidase and other hepatic function tests were normal. The erythrocyte sedimentation rate (ESR) was raised in five (45%) of 11 patients, and was more than 100 mm in 1 h in two patients. There was a consistent five-fold to ten-fold increase in C reactive protein in four (40%) of
ten patients, and inflammatory anaemia was seen in three (25%) of 12 patients. Leucocytosis was seen in two (17%) of 12 patients, of whom one had an increased monocyte count (1·3⫻109/L). Decreased lymphocyte counts of 0·53⫻109/L and 0·93⫻109/L were seen in two (17%) of 12 patients. The presence of various circulating autoantibodies was searched for in nine patients and were present in the patient with systemic lupus erythematosus (antinuclear, anticardiolipin, anti-SSB antibodies) and the patient with rheumatoid arthritis (rheumatoid factor). Muscle biopsy (figure 1) showed conspicuous infiltration of connective-tissue structures by densely packed large and grossly rounded cells, with central, round, and often nucleolated nuclei, and clear slightly basophilic cytoplasm with fine PAS-positive granules and, rarely, lipid droplets. On immunocytochemistry, these cells were negative for desmin and smooth-muscle action. They were positive for the histiocyte marker CD68, and negative for the Langerhans-cell marker CD1a and for S100 protein, and were, therefore, recognised as macrophages. Necrosis, epithelioid cells, giant cells, and intracytoplasmic inclusions, such as Michaelis-Gutmann bodies, phagocytised blood cells, and foreign bodies, were not seen. Ziehl-Neelsen, auramine, and gram stainings were negative. Mitotic figures were absent. Extension of the macrophage infiltrate were focal or multifocal and were seen in deep subcutaneous fat, epimysium, perimysium, or endomysium (table 4). Occasionally, the infiltrate appeared as a thick epimysial coating. In the endomysium, the infiltrate remained confined to
Figure 2: Muscle biopsy on light microscope A=massive perifascicular infiltration by macrophages surrounding a lymphocytic follicle (patient 10, haematoxylin and eosin, ⫻70); B=sheet of macrophages infiltrated from epimysium to perimysium (patient two, haematoxylin and eosin, ⫻200); C=slightly basophilic non-epithelioid cells corresponding to macrophages infiltrated between muscle fibres (patient 10, haematoxylin and eosin, ⫻500); D=centripetal infiltration of perimysium and perifascicular endomysium by macrophages expressing CD68 (patient 5, alkaline phosphatase antialkaline phosphatase assay, ⫻300); E=focal endomysial accumulation of macrophages filled with a PAS-positive material (patient four, PAS, ⫻180); F=sparse CD8-T-cells intermingled with macrophages in endomysium (patient 4, APAAP, ⫻200).
THE LANCET • Vol 352 • August 1, 1998
Figure 3: Muscle biopsy on electron microscope A=macrophage cytoplasmic process showing aggregates of dense spicules, some of which are bounded by membrane (patient 10, ⫻32 000); B=ultrastructural view of dense intracytoplasmic inclusion bodies in macrophage: spicules are about 1·2 nm larger and 60 nm to 120 nm long (patient 10, ⫻80 000).
349
ARTICLES
peripheral regions of the muscle fascicle and tended to smother muscle fibres. A contiguous and seemingly centripetal extension of the infiltrate from epimysium to perimysium, from perimysium to endomysium, or from perifascicular to deeper endomysium was common. In some cases, the infiltrate was perivascular or associated with mild to moderate fibrous thickening of the connective tissue. Lymphocytic infiltrates were seen in most cases (table 4). The infiltrates were generally mild, appearing as isolated lymphocytes intermingled with macrophages or as small perivascular cuffs, close to or remote from the macrophage infiltrate. In one case, a lymphocytic follicle was seen in epimysium. Lymphocytes were mainly CD8 T cells. CD4 T cells were rare and CD 20 B cells and plasma cells were virtually absent. Myonecrosis, perifascicular atrophy, and myopathic changes were inconspicuous or absent. Major histocompatibility complex class II antigen was expressed by macrophages. Unlike polymyositis, major histocompatibility complex class I antigen was not expressed by muscle fibres. Unlike dermatomyositis, there were no significant deposits of membrane-attack complex (C5b9) in endomysial capillaries. In the four patients in whom the infiltrate was present in the epoxy-embedded sample, electron microscopy showed histiocytic cells without Birbeck granules but with intracytoplasmic aggregates of fine needle-shaped randomly oriented dense structures forming clusters, often bounded by distinct limiting membranes (figure 2). The material showed some similarities with apatite crystals, but extracellular deposits were not seen, and Von Kossa and Alizarin red S stainings were negative. No structures reminiscent of bacteria or viruses were seen in cells or extracellular spaces. Routine PCR-based tests for Tropheryma whippelii, Mycobacterium tuberculosis complex, and Mycobacterium avium complex from muscle biopsy samples of five patients were negative. PAS-positive macrophages were not seen in sites other than muscle. The initial diagnosis of Whipple’s disease was considered in many patients, but it could not be substantiated by appropriate biopsies of duodenum (seven patients) and other parts of the digestive tract (eight patients). Repeat duodenal biopsy in one patient remained negative. PCR-based detection of T whippelii from jejunal tissue in one case was negative. Tuberculin skin test was negative in one patient, and 7 mm or less in two patients. Cultures of bronchial aspirates and urine were negative for M tuberculosis (four patients). Syphilis tests were negative (four patients). Serological tests for HIV-1 and HIV-2 (eight patients), human T-cell lymphoma virus types 1 and 2 (two patients), hepatitis A, B, and C viruses (six patients), Epstein-Barr virus (seven patients), cytomegalovirus (six patients), Toxoplasma gondii (four patients), Trichinella spiralis (two patients), Coxiella burnetti (two patients), Rickettsia conorii (two patients), Q fever (one patient), and Bartonella henselae (one patient) infections were negative or showed residual antibodies. Other normal, negative, or non-contributory laboratory tests or findings included chest radiography (nine patients), serum angiotensin-converting enzyme (one patient), bronchoalveolar lavage (two patients), pulmonary function tests (two patients), effort electrocardiogram (two patients), cardiac echogram (one patient), imaging of the abdominal cavity (four patients), 350
cerebrospinal fluid examination (two patients), imaging of the central nervous system (four patients), thyroid tests (nine patients), cortisolaemia (one patient), and detection of various circulating markers of cancer (one patient). Four patients received no treatment and ten received various combinations of anti-inflammatory drugs and antibiotics (table 1). Four patients were treated by different antibiotics, including co-trimoxazole (three patients), standard antituberculous antibiotic therapy (two patients), chloramphenicol (one patient), minocycline (one patient), azythromycin (one patient), and ofloxacin (one patient). Complete improvement with antibiotics alone was seen in one patient (patient 8, who received co-trimoxazole and antituberculous drugs pyrazinamide, ethambutol, isoniazid, and rifampicin). Two patients partially improved under antibiotic therapy (patient 1 received co-trimoxazole plus chloramphenicol; patient 7 received minocycline) and completely improved with subsequent administration of steroids. No improvement was seen in patient six with trimethoprim, azythromycin, or ofloxacin, each used alone. Non-steroidal anti-inflammatory drugs were ineffective. Steroids were used at various doses in eight of 14 patients and resulted in partial (three) or complete recovery (five) in a few days or weeks.
Discussion In several patients seen since 1993 with varied combinations of arthromyalgias, myopathic weakness, increased creatine kinase concentrations, and raised erythrocyte sedimentation rate, muscle biopsy had shown a unique pattern of conspicuous accumulations of large macrophages with fine PAS-positive granular cytoplasm in the epimysium, perimysium, or endomysium, associated with mild CD8 T-cell infiltrates, and minimum muscle fibre damage. PAS-positive macrophages were not detected elsewhere other than in muscle, fascia, and the deep subcutaneous tissue, although symptoms suggesting possible involvement of other organs were present in a few patients. Routine investigations did not show the cause of the disorder. Patients responded well to steroids, with or without antibiotics. Unlike sarcoid granulomatas, macrophages infiltrating muscle connective tissue did not have an epithelioid appearance, did not form giant cells, and were not organised in nodules. Macrophagic myofasciitis was, therefore, easily distinguishable from sarcoid myopathy,3 and was clearly different from the fasciitis-panniculitis syndromes, such as Shulman’s disease, toxic-oil syndrome, and eosinophilia-myalgia syndrome, in which macrophages epithelioid cells, and giant cells may accompany chronic inflammation, fibrosis, and vascular changes but never predominate in the infiltrate.10 With the exception of sarcoid-like granulomatous myopathies, conspicuous accumulations of histiocytes in skeletal muscle are extremely rare. Histiocytes may derive from Langerhans or dendritic cells, or, as in our patients, from monocytes or macrophages.11 Other than in sarcoidosis and sarcoid-like disorders12 “reactive” accumulations of macrophages occur in storage diseases related to inborn errors of metabolism,13 benign proliferations of macrophages known as non-Langerhans histiocytoses,11,14 and infectious diseases associated with a defect in macrophage handling of bacteria.15
THE LANCET • Vol 352 • August 1, 1998
ARTICLES
In our patients, the nature of the accumulated material in macrophages remained undetermined, and, therefore, a storage disease should be considered. However, the histories of our patients gave no support for an inherited metabolic disease, or for accumulation of an organic material of iatrogenic origin. Light-microscopy and electron microscopy characteristics of the accumulated material were not those of the ceroid pigment found in sea-blue histiocytes, and the disorder seemed to be an inflammatory disease rather than a metabolic disorder. We are aware of no storage disease that selectively involves macrophages and produces symptoms in muscles. Non-Langerhans histiocytoses may present with 11 different clinicopathological patterns.11 One of these patterns, multicentric reticulohistiocytosis, may be associated with ill-defined muscle involvement.16 Multicentric reticulohistiocytosis, however, cannot be confused with macrophagic myofasciitis since nodular skin lesions, severe destructive arthropathy, and abundant multinucleated giant-cell formation are the hallmarks of multicentric reticulohistiocytosis.11,14 Infectious diseases associated with macrophage dysfunction are generally characterised by the presence of PAS-positive material within macrophages, as in macrophagic myofasciitis. These diseases include malakoplakia,17 Whipple’s disease,18 and M avium complex infection.19,20 Malakoplakia is a rare chronic inflammatory disorder associated with chronic infections, mainly by coliform bacilli, characterised by collections of macrophages in the genitourinary tract and, less frequently, in other sites.17 This disorder is not thought to cause muscle disease.3 Histopathologically, intracellular and extracellular calcified bodies with a laminar structure, known as Michaelis-Gutmann bodies, are the pathognomonic feature of the disease.17 Such inclusions were not seen in our patients, and, therefore, macrophagic myofasciitis should not be diagnosed as muscle malakoplakia. We initially considered the diagnosis of Whipple’s disease for many of our patients because of the report of one case of progressive myopathy associated with this disease,21 and the rare occurrence of Whipple’s disease without intestinal involvement (estimated at ten of 760 patients18). In the reported patient with myopathy, muscle biopsy showed macrophages containing typical PAS-positive sickle-shaped particles in the perimysium.21 Intracytoplasmic membranous bodies and bacillary bodies were found on electronmicroscopy in macrophages in muscle and jejunum. Oral antibiotics improved the myopathy. We are aware of only one other reference to specific myopathy during Whipple’s disease.22 None of the seven myopathologists who were coinvestigators in our study has seen such a case. A clinical review of 52 French patients with Whipple’s disease did not mention muscular symptoms, even nonspecific symptoms.23 It is unlikely that our patients had Whipple’s disease since they had no digestive symptoms, no PAS-positive macrophages in the lamina propria of the digestive tract, of T whippelii in the jejunum or muscle, and no ultrastructural evidence of bacillary bodies within or in the vicinity of macrophage infiltrates in muscle. A few patients infected by non-tuberculous mycobacterial infections may develop a pronounced
THE LANCET • Vol 352 • August 1, 1998
systemic histiocytic proliferation that mainly affects lymphoid organs and the gastrointestinal tract.19,20 Specific muscle involvement by the M avium complex has rarely been observed in patients who have died from AIDS.24 The histological picture has not generally indicated severe necrosis or giant-cell reaction, but has been characterised by the presence of abundant intracellular acid-fast bacilli, easily detectable by special stains and electron microscopy.19,20,24 Ziehl-Neelsen and auramin stains as well as PCR detection of M avium complex were negative in our patients, thus precluding the diagnosis of M avium complex infection. Finally, macrophagic myofasciitis does not correspond to any of the previously described histiocytoses or reactive macrophage accumulations. At present the cause of this unique disorder is unknown. Most emerging or re-emerging diseases, however, are or will probably be caused by infectious agents.1 The spectrum of recognised mycobacterial diseases, has, for example, increased in the past few years.25 Indeed, there are some indirect arguments for an infectious cause of macrophagic myofasciitis. Exposure to tuberculosis was noted in two patients, of whom one improved completely under antituberculous quadritherapy combined with cotrimoxazole. Like patients with malakoplakia, Whipple’s disease, or histiocytosis associated with M avium complex, some of our patients had an altered immunological status. Four patients had concomitant systemic disorders—malignant tumour in one, and autoimmune disease in three. One patient had monocytosis, and two had a persistent lymphocytopenia consistent with an altered cell-mediated immunity. A striking feature was that five patients took chloroquine or hydroxychloroquine. Impaired antimycobacterial activity of macrophages is linked to impaired secretion of tumour necrosis factor ␣ and interleukins 1-␣ and 1-, 6, and 12 by macrophages and impaired secretion of interferon-␥ by T lymphocytes.15,26–28 Interestingly, hydroxychloroquine inhibits secretion of interleukins 1-␣ and 6 by macrophages.29 This action may interfere with the complex cytokine interplay regulating antimycobacterial activity and may be harmful; similar effects may be seen with other drugs that affect host defence against mycobacteria.27 Further work is needed to identify the cause of macrophagic myofasciitis, but the disorder is potentially treatable. Contributors R K Gherardi, L Belec, and M Fardeau designed the study. P Cherin and F-J Authier were responsible for clinical investigations. R K Gherardi, M Coquet, F-J Authier, D Figarella-Branger, J-M Mussini, J-F Pellissier, and M Fardeau were responsible for histopathological invstigations. L Belec did molecular microbiological assessments. R K Gherardi and F-J Authier wrote the paper. All authors read and edited the manuscript.
Acknowledgments We thank D Bernard, J M Berthelot, C Bihel-Schleich, R Carlioz, J-Y Colin, B Eymard, F Gray, P Heine, S Herson, J M Labrousse, J-P Leroy, X Pennaforte, D Ranoux, D Sereni, A Schaeffer, J A Urtizberea, F Vaylet, and P-J Weller, who referred or contributed to the assessment of the patients. This work was supported by the Association Française contre les Myopathies (AFM).
References 1
2
Schwartz DA. Emerging and reemerging infections: progress and challenges in the subspecialty of infectious disease pathology. Arch Pathol Lab Med 1997; 121: 776–84. Dalakas MC. Polymyositis, dermatomyositis and inclusion body myositis. N Engl J Med 1991; 325: 1487–98.
351
ARTICLES 3
4
5
6
7 8
9
10
11 12
13 14
15
16
17
Banker BQ. Other inflammatory myopathies. In: Engel AG, Franzini-Armstrong, eds. Myology, 2nd edn, vol 2. New York: McGraw-Hill, 1993: 1461–86. Alonzo-Ruiz A, Zea-Mendoza AC, Salazar-Vallinas JM, Rocamora-Ripoll A, Beltran-Guttierrez D. Toxic oil syndrome: a syndrome with features overlapping those of various forms of scleroderma. Semin Arthritis Rheum 1986; 15: 200–12. Kaufman LD, Seidman RJ, Gruber BL. L-tryptophan-associated eosinophilic perimyositis, neuritis and fasciitis: a clinicopathogic and laboratory study of 25 patients. Medicine 1990; 69: 187–99. Mayeno AN, Benson LM, Naylor S, Colberg-Beers M, Puchalski JT, Gleich GJ. Biotransformation of 3-(phenylamino)1,2-propanediol to 3-(phenylamino)alanine: a chemical link between toxic oil syndrome and eosinophilia-myalgia syndrome. Chem Res Toxicol 1995; 8: 911–16. Gherardi R. Skeletal muscle involvement in HIV-infected patients. Neuropathol Appl Neurobiol 1994; 20: 232–37. Dalakas MC, Illa I, Pezeshkpour GH, Laukaitis JP, Cohen B, Griffin JL. Mitochondrial myopathy caused by long-term zidovudine therapy. N Engl J Med 1990; 322: 1098–05. Mhiri C, Baudrimont M, Bonne G, et al. Zidovudine myopathy: a distinctive disorder associated with mitochondrial dysfunction. Ann Neurol 1991; 29: 606–14. Naschitz JE, Boss JH, Misselevitch I, Yeshurun D, Rosner I. The fasciitis-panniculitis syndromes: clinical and pathologic features. Medicine 1996; 75: 6–16. Snow JL, Daniel Su WP. Histiocytic disorders. J Am Acad Dermatol 1995; 33: 111–16. Friedlander JS, Weatherall DJ, Ledingham JGG. A chronic granulomatous syndrome of unknown origin. Medicine 1990; 69: 325–31. Cline MJ. Histiocytes and histiocytosis. Blood 1994; 84: 2840–53. Zellger BWH, Sidoroff A, Orchard G, Cerio R. Non-Langerhans cells histiocytosis: a new unifying concept. Am J Dermatopathol 1996; 18: 490–504. Marth T, Neurath M, Cuccherini BA, Strober W. Defects of monocyte interleukin 12 production and humoral immunity in Whipple’s disease. Gastroenterology 1997; 113: 442–48. Gao IK, Goronzy JJ, Weyand CM. Ten year follow-up of a patient with multicentric reticulohistiocytosis associated with myopathy. Scand J Rheumatol 1990; 19: 437–41. Lack EE. Malakoplakia. In: Connor DH, Chandler FW, Schwartz DA, Manz HJ, Lack EE, eds. Pathology of infectious
352
diseases, vol 2. Stamford, USA: Appleton and Lange, 1997: 1647–54. 18 Dobbins WO III. Whipple’s disease. In: Connor DH, Chandler FW, Schwartz DA, Manz HJ, Lack EE, eds. Pathology of infectious diseases, vol 1. Stamford, USA: Appleton and Lange, 1997: 897–903. 19 Miranda D, Vuletin JC, Kauffman SL. Disseminated histiocytosis and intestinal malakoplakia: occurrence due to Mycobacterium intracellulare infection. Arch Pathol Lab Med 1979; 103: 302–05. 20 Gillin JS, Urmacher C, West R, Shike M. Disseminated Mycobacterium avium-intracellulare infection in acquired immunodeficiency syndrome mimicking Whipple’s disease. Gastroenterology 1983; 85: 1187–91. 21 Swash M, Schwartz MS, Vandenburg MJ, Pollock DJ. Myopathy in Whipple’s disease. Gut 1977; 18: 800. 22 Banker BQ, Engel AG. Basic reactions of muscle. In: Engel AG, Franzini-Armstrong, eds. Myology, 2nd edn, vol 1. New York: McGraw-Hill, 1993: 832–88. 23 Vital Durand D, Lecomte C, Cathébras P, Rousset H, Godeau P, SNFMI research group on Whipple’s disease. Whipple disease: clinical review of 52 cases. Medicine 1997; 76: 170–84. 24 Wrzolek MA, Sher JH, Kozlovski PB, Rao C. Skeletal muscle pathology in AIDS: an autopsy study. Muscle Nerve 1990; 13: 508–15. 25 Watt B. Lesser known mycobacteria. J Clin Pathol 1995; 48: 701–05. 26 Michelini-Norris MB, Blanchard DK, Pearson CA, Djeu JY. Differential release of interleukin (IL)-1 beta, and IL-6 from normal human monocytes stimulated with a virulent and an avitulent isogenic variant of Mycobacterium avium-intracellulare complex. J Infect Dis 1992; 165: 702–09. 27 Sathe SS, Sarai A, Tsigler D, Nedunchezian D. Pentoxifylline aggravates impairment in tumor necrosis factor-alpha secretion and increases mycobacterial load in macrophages from AIDS patients with disseminated Mycobacterioum avium-intracellulare complex infection. J Infect Dis 1994; 170: 484–87. 28 Frucht DM, Holland SM. Defective monocyte costimulation for IFN-␥ production in familial disseminated Mycobacterium avium complex infection: abnormal IL-12 regulation. J Immunol 1996; 157: 41–16. 29 Sperber K, Quraishi H, Kalb TH, Panja A, Stecher V, Mayer L. Selective regulation of cytokine secretion by hydroxychloroquine: inhibition of interleukin 1 alpha (IL-1 alpha) and IL-6 in human monocytes and T-cells. J Rheumatol 1993; 2): 803–08.
THE LANCET • Vol 352 • August 1, 1998
Macrophagic myofasciitis: an emerging entity R K Gherardi, M Coquet, P Chérin, F-J Authier, P Laforêt, L Bélec, D Figarella-Branger, J-M Mussini, J-F Pellissier, M Fardeau, for the Groupe d’Études et Recherche sur les Maladies Musculaires Acquises et Dysimmunitaires (GERMMAD) de l’Association Française contre les Myopathies (AFM)
Summary Background An unusual inflammatory myopathy characterised by an infiltration of non-epithelioid histiocytic cells has been recorded with increasing frequency in the past 5 years in France. We reassessed some of these cases. Methods We did a retrospective analysis of 18 such cases seen in five myopathology centres between May, 1993, and December, 1997. The myopathological changes were reassessed at a clinopathology seminar. Findings Detailed clinical information was available for 14 patients. The main presumptive diagnoses were polymyositis and polymyalgia rheumatica. Symptoms included myalgias in 12 patients, arthralgias in nine, muscle weakness in six, pronounced asthenia in five, and fever in four. Abnormal laboratory findings were occasionally observed, and included raised creatine kinase concentrations, increased erythrocyte sedimentation rate, and myopathic electromyography. Muscle biopsy showed infiltration of the subcutaneous tissue, epimysium, perimysium, and perifascicular endomysium by sheets of large macrophages, with a finely granular PAS-positive content. Also present were occasional CD8 T cells, and inconspicuous muscle-fibre damage. Epithelioid and giant cells, necrosis, and mitotic figures were not seen. The images were easily distinguishable from sarcoid myopathy and fasciitis-panniculitis syndromes. Whipple’s disease, Mycobacterium avium intracellulare infection, and malakoplakia could not be confirmed. Ten patients were treated with various combinations of steroids and antibiotics; symptoms improved in eight patients, and stabilised in two. Interpretation A new inflammatory muscle disorder of unknown cause, characterised by a distinctive pathological pattern of macrophagic myofasciitis, is emerging in France.
Lancet 1998; 352: 347–52 Groupe d’Etude et de Recherche sur le Muscle et le Nerf (GERMEN), EA 2347, Université Paris XII-Val de Marne, Département de Pathologie, Hôpital Henri Mondor, F-94010 Créteil, France (Prof R K Gherardi MD, F-J Authier MD); Unité de Myopathologie, Départment d’Anatomie Pathologique, Centre Hospitalier Universitaire de Bordeaux, Hôpital Pellegrin, Bordeaux (M Coquet MD); Service de Médecine Interne et Institut de Myologie-INSERM U153 Groupe Hospitalier Pitié-Salpêtrière, Paris (P Cherin MD, P Laforêt MD, Prof M Fardeau MD); Service de Microbiologie, Hôpital Broussais, Paris (L Belec MD); Laboratoire de Biopathologie Nerveuse et Musculaire (JE 2053, Université Aix-Marseille II), Faculté de Médecine, Marseille (Prof D Figarella-Branger MD, Prof J-F Pellissier MD); and Laboratoire d’Anatomie Pathologique A, Bâtiment Jean Monnet, CHR 44000 Nantes (J-M Mussini MD) Correspondence to: Prof R K Gherardi (e-mail: [email protected])
THE LANCET • Vol 352 • August 1, 1998
Introduction Emerging diseases are challenging clinicians and pathologists with increasing frequency.1 Careful analysis of muscle biopsy samples have provided valuable evidence of the pathogenesis of idiopathic inflammatory myopathies,2 and myopathological alterations allow classification of these disorders into three main categories: dermatomyositis, polymyositis, and inclusionbody myositis. Other inflammatory myopathies3 include sarcoid and sarcoid-like granulomatous myopathies, and various autoimmune, infectious, or toxic diseases. The Spanish toxic-oil syndrome,4 for example, and the adulterated L-tryptophan-associated eosinophiliamyalgia syndrome5 are closely-related disorders6 associated with fasciitis, myositis, and systemic involvement, that appeared in 1981 and 1989, respectively. Other emerging inflammatory muscle diseases are those associated with HIV-1 infection, the incidence of which rose sharply after 1987,7 because of zidovudine myopathy, a toxic mitochondrial myopathy associated with myositis.8,9 We report the emergence of a new type of inflammatory myopathy of unkown origin, characterised by a unique myopathological pattern of infiltration of the epimysium, perimysium, and perifascicular endomysium by non-epithelioid periodic-acid-schiff (PAS)-positive cells of macrophage lineage. The disorder is easily distinguishable from sarcoid myopathy and fasciitis-panniculitis syndromes, was first seen in 1993, and has increased in frequency since that time in five French myopathological centres.
Methods From May, 1993, to December, 1997, an unusual infiltration of skeletal-muscle connective-tissue structures by non-epithelioid histiocytic cells was seen with increasing frequency in the myopathology centres of Bordeaux (three cases), Marseilles (three cases), Nantes (one case), and Paris (University Hospital Pitié-Salpêtriere, four cases, University Hospital Henri Mondor, Créteil, seven cases; figure 1). The yearly number of muscle biopsies examined in these myopathology centres is about 1200. All samples were collected as part of the routine tests in these laboratories. The patients were adults who had been referred with the presumptive diagnosis of polymyositis (ten cases), polymyalgia rheumatica (four cases), muscle dystrophy (one
Figure 1: Cases of macrophagic myofasciitis
347
ARTICLES
Case number
Sex/age (yr)
Duration of symptoms (mo)
1
F/55
6
2 3 4 5
M/38 M/48 F/70 F/56
11 36 3 24
6 7 8 9 10 11 12 13
M/35 F/37 M/61 M/53 F/29 M/25 F/30 F/52
4 14 12 48 24 12 6 7
14
F/29
3
Musculoskeletal symptoms
Other symptoms
EMG findings
CK
Myalgias, arthralgias, weakness, tenderness Myalgias, arthralgias, weakness Myalgias, arthralgias Myalgias, arthralgias Myalgias, arthralgias, weakness, tenderness Myalgias, arthralgias Myalgias, arthralgias Myalgias, arthralgias Weakness Myalgias Myalgias Myalgias, arthralgias, tenderness Myalgias, arthralgias, weakness
Ankle oedema
Myopathic
Normal
55
Remission (on treatment*)
Chest pain Ankle oedema None Headaches, fever, fatigue Fatigue None Cough, fever Dyspnoea, fatigue None Dyspnoea, fever, fatigue Dyspnoea, fever, fatigue Cough, peripheral neuropathy None
Not done Normal Normal Myopathic
⫻4 ⫻3 Normal ⫻100
12 10 110 115
Remission (on treatment†) Spontaneous remission Remission (on treatment†) Remission (on treatment†)
Normal Normal Not done Normal Myopathic Normal Myopathic Neuropathic
⫻1·5 Normal Normal Normal ⫻5 ⫻3 Normal Normal
Not done 10 70 Not done 4 8 26 6·5
Stabilisation (on treatment*) Remission (on treatment†) Remission (on treatment‡) Slow progression Remission (on treatment†) Slow progression Remission (on treatment†) Stabilisation (on treatment†)
Neuropathic
Normal
Not done
Slow progression
Weakness
ESR (mm/h)
Outcome
EMG=electromyogram; CK=creatine kinase; ESR=erythrocyte sedimentation rate. *Antibiotics combined with steroids. †Steroids alone. ‡Antibiotics alone.
Table 1: Clinical and biological findings in patients with macrophagic myofasciitis case), or mitochondrial cytopathy (three cases). Detailed clinical information was available for 14 of the 18 patients. The deltoid-muscle biopsy samples had been conventionally divided into three portions, frozen, and embedded in paraffin and epoxy, and had been processed in the standard way. Myopathological alterations were systematically reassessed by five myopathologists (RG, MC, J-MM, J-FP, MF) in September, 1997, during a clinicopathological seminar of the Groupe d’Etudes et Recherche sur les Maladies Musculaires Acquises et Dysimmunitaires (GERMMAD), a branch of the Association Française contre les Myopathies. Conclusions of the seminar were transmitted to the Réseau National de Santé Publique France, and to the Centers for Disease Control, USA.
Results Characteristics of patients are listed in table 1. 12 patients were white, one was West Indian, and one was Algerian. Four patients had stayed for long times in tropical countries (three in Africa, one in the West Indies), and four were working in a hospital. Four patients were taking no medication at the time of first muscular symptoms, and four were on various drugs not known to cause myopathy, four were on chloroquine to prevent malaria, one took hydroxychloroquine as an immunomodulator, and one was on lipid-lowering drugs that were withdrawn, but symptoms did not improve. Two patients had had tuberculosis and one had been cured of sarcoidosis. Medical history (from 1 year before first muscle symptoms) included a connective-tissue disease in three patients (systemic lupus erythematosus, rheumatoid arthritis, and Hashimoto’s thyroiditis), myelodysplasia in one, and uvula carcinoma treated by irradiation at time of first muscular symptoms in one. Clinical features and abnormal laboratory findings are summarised in tables 2 and 3. All patients had had Symptom
Number of patients (n=14)
Myalgias Arthralgias Muscle weakness Fatigue Fever Muscle tenderness Dyspnoea Acroparaesthesiae in hands Ankle oedema Dry cough Epigastric pain Chest pain Dysphagia Headache
12 9 6 5 4 4 3 2 2 2 2 1 1 1
Table 2: Symptoms of patients with macrophagic myofasciitis
348
muscular symptoms for 3–48 months before biopsy. Mild to moderate chronic myalgias were the most common manifestation, occurring in 12 (86%) of 14 patients. In two patients, muscle pain was localised in upper limbs (girdle, arms, and forearms), in four patients it was localised in upper limbs and thighs, and in six patients myalgias were diffuse. In three patients, myalgias started or increased on exertion. Arthralgias were present in nine (64%) of 14 patients, with large joints involved in all nine patients (shoulders, elbows, wrists, lumbosacral spine, or hips) and finger joints in three. Six of 14 patients (43%) complained of mild to moderate muscle weakness, with proximal muscles of all arms and legs affected in four patients, and selective involvement of lower limbs in two. Six (43%) of 14 patients had general symptoms, including marked asthenia (five) and fever ranging from 38ºC to 39ºC (four). Digestive symptoms other than epigastric pain were absent at presentation. One patient developed diarrhoea when on long-term antibiotic therapy. Physical examination showed no muscle wasting. Muscle tenderness was present in four (29%) of 14 patients, and muscle induration in two (14%). Skin changes, including peau d’orange induration, were not seen. Except for nocturnal paresthaesias in hands and a bilateral Tinel’s sign noted in two (14%) of 14 patients, there was no peripheral neuropathy. There was neither lymphadenopathy nor hepatosplenomegaly. Electromyography was myopathic in four (33%) of 12 patients and equivocal in two (17%) of 12. Study of nerve-conduction velocity showed evidence of sensoryperipheral-nerve involvement in two patients. Serum Test
Number of patients positive
Number of patients tested
Creatine kinase >200 IU/L Aldolase >8 IU/L AST/ALT >35 IU/L Erythrocyte sedimentation rate >40 mm/h C-reactive protein >0·8 mg/dL Haemoglobin <12 g/L Lymphocytes 1·0⫻109/L Monocytes 0·5⫻109/L Myopathic electromyography Abnormal muscle magnetic-resonance imaging
6 3 2 5 4 3 2 1 4 1
14 7 11 11 10 10 10 10 12 4
AST/ALT=aspartate aminotransferase/alanine aminotransferase. Autoantibodies detected in patient with known systemic lupus erythematosus, and rheumatoid factor detected in patient with rheumatoid arthritis were excluded.
Table 3: Abnormal laboratory findings in patients with macrophagic myofasciitis
THE LANCET • Vol 352 • August 1, 1998
ARTICLES
Case number
1 2 3 4 5 6 7 9 10 11 12 13 14
Macrophagic infiltrate
T-cell infiltrate
Epimysium
Perimysium
Endomysium
Giant cells
PAS-positive
Sparse
Perivascular
Myonecrosis
+++ +++ +++ ⫺ +++ +++ ++ ⫺ ++ ++ +++ +++ +++
+++ ++ ⫺ + +++ +++ ++ ++ +++ ++ ++ + ++
+++ + ⫺ +++ + ++ + + ++ + + ⫺ ++
⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺ ⫺
Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
+ + ⫺ + ⫺ + + ⫺ + + ⫺ + +
⫺ + + + ++ ⫺ + ⫺ +++ + + ⫺ +
+/⫺ ⫺ +/⫺ ⫺ ⫺ ⫺ +/⫺ ⫺ ⫺ ⫺ ⫺ +/⫺ ⫺
⫺=absent; +/⫺=almost completely absent; +=mild; ++=moderate; +++=severe.
Table 4: Myopathological findings in patients with macrophagic myofasciitis
creatine-kinase concentration was raised in six (43%) of 14 patients, with increase of 1·5–5·0 times the normal in five patients, and more than a ten-fold increase in one. Aldolase concentration was slightly raised in three (43%) of seven patients, including one who had a normal creatine kinase concentration. Aspartateaminotransferase and alanine-aminotransferase were increased consistently with the raised creatine kinase concentrations in two (18%) of 11 patients. ␥-glutamyl transpeptidase and other hepatic function tests were normal. The erythrocyte sedimentation rate (ESR) was raised in five (45%) of 11 patients, and was more than 100 mm in 1 h in two patients. There was a consistent five-fold to ten-fold increase in C reactive protein in four (40%) of
ten patients, and inflammatory anaemia was seen in three (25%) of 12 patients. Leucocytosis was seen in two (17%) of 12 patients, of whom one had an increased monocyte count (1·3⫻109/L). Decreased lymphocyte counts of 0·53⫻109/L and 0·93⫻109/L were seen in two (17%) of 12 patients. The presence of various circulating autoantibodies was searched for in nine patients and were present in the patient with systemic lupus erythematosus (antinuclear, anticardiolipin, anti-SSB antibodies) and the patient with rheumatoid arthritis (rheumatoid factor). Muscle biopsy (figure 1) showed conspicuous infiltration of connective-tissue structures by densely packed large and grossly rounded cells, with central, round, and often nucleolated nuclei, and clear slightly basophilic cytoplasm with fine PAS-positive granules and, rarely, lipid droplets. On immunocytochemistry, these cells were negative for desmin and smooth-muscle action. They were positive for the histiocyte marker CD68, and negative for the Langerhans-cell marker CD1a and for S100 protein, and were, therefore, recognised as macrophages. Necrosis, epithelioid cells, giant cells, and intracytoplasmic inclusions, such as Michaelis-Gutmann bodies, phagocytised blood cells, and foreign bodies, were not seen. Ziehl-Neelsen, auramine, and gram stainings were negative. Mitotic figures were absent. Extension of the macrophage infiltrate were focal or multifocal and were seen in deep subcutaneous fat, epimysium, perimysium, or endomysium (table 4). Occasionally, the infiltrate appeared as a thick epimysial coating. In the endomysium, the infiltrate remained confined to
Figure 2: Muscle biopsy on light microscope A=massive perifascicular infiltration by macrophages surrounding a lymphocytic follicle (patient 10, haematoxylin and eosin, ⫻70); B=sheet of macrophages infiltrated from epimysium to perimysium (patient two, haematoxylin and eosin, ⫻200); C=slightly basophilic non-epithelioid cells corresponding to macrophages infiltrated between muscle fibres (patient 10, haematoxylin and eosin, ⫻500); D=centripetal infiltration of perimysium and perifascicular endomysium by macrophages expressing CD68 (patient 5, alkaline phosphatase antialkaline phosphatase assay, ⫻300); E=focal endomysial accumulation of macrophages filled with a PAS-positive material (patient four, PAS, ⫻180); F=sparse CD8-T-cells intermingled with macrophages in endomysium (patient 4, APAAP, ⫻200).
THE LANCET • Vol 352 • August 1, 1998
Figure 3: Muscle biopsy on electron microscope A=macrophage cytoplasmic process showing aggregates of dense spicules, some of which are bounded by membrane (patient 10, ⫻32 000); B=ultrastructural view of dense intracytoplasmic inclusion bodies in macrophage: spicules are about 1·2 nm larger and 60 nm to 120 nm long (patient 10, ⫻80 000).
349
ARTICLES
peripheral regions of the muscle fascicle and tended to smother muscle fibres. A contiguous and seemingly centripetal extension of the infiltrate from epimysium to perimysium, from perimysium to endomysium, or from perifascicular to deeper endomysium was common. In some cases, the infiltrate was perivascular or associated with mild to moderate fibrous thickening of the connective tissue. Lymphocytic infiltrates were seen in most cases (table 4). The infiltrates were generally mild, appearing as isolated lymphocytes intermingled with macrophages or as small perivascular cuffs, close to or remote from the macrophage infiltrate. In one case, a lymphocytic follicle was seen in epimysium. Lymphocytes were mainly CD8 T cells. CD4 T cells were rare and CD 20 B cells and plasma cells were virtually absent. Myonecrosis, perifascicular atrophy, and myopathic changes were inconspicuous or absent. Major histocompatibility complex class II antigen was expressed by macrophages. Unlike polymyositis, major histocompatibility complex class I antigen was not expressed by muscle fibres. Unlike dermatomyositis, there were no significant deposits of membrane-attack complex (C5b9) in endomysial capillaries. In the four patients in whom the infiltrate was present in the epoxy-embedded sample, electron microscopy showed histiocytic cells without Birbeck granules but with intracytoplasmic aggregates of fine needle-shaped randomly oriented dense structures forming clusters, often bounded by distinct limiting membranes (figure 2). The material showed some similarities with apatite crystals, but extracellular deposits were not seen, and Von Kossa and Alizarin red S stainings were negative. No structures reminiscent of bacteria or viruses were seen in cells or extracellular spaces. Routine PCR-based tests for Tropheryma whippelii, Mycobacterium tuberculosis complex, and Mycobacterium avium complex from muscle biopsy samples of five patients were negative. PAS-positive macrophages were not seen in sites other than muscle. The initial diagnosis of Whipple’s disease was considered in many patients, but it could not be substantiated by appropriate biopsies of duodenum (seven patients) and other parts of the digestive tract (eight patients). Repeat duodenal biopsy in one patient remained negative. PCR-based detection of T whippelii from jejunal tissue in one case was negative. Tuberculin skin test was negative in one patient, and 7 mm or less in two patients. Cultures of bronchial aspirates and urine were negative for M tuberculosis (four patients). Syphilis tests were negative (four patients). Serological tests for HIV-1 and HIV-2 (eight patients), human T-cell lymphoma virus types 1 and 2 (two patients), hepatitis A, B, and C viruses (six patients), Epstein-Barr virus (seven patients), cytomegalovirus (six patients), Toxoplasma gondii (four patients), Trichinella spiralis (two patients), Coxiella burnetti (two patients), Rickettsia conorii (two patients), Q fever (one patient), and Bartonella henselae (one patient) infections were negative or showed residual antibodies. Other normal, negative, or non-contributory laboratory tests or findings included chest radiography (nine patients), serum angiotensin-converting enzyme (one patient), bronchoalveolar lavage (two patients), pulmonary function tests (two patients), effort electrocardiogram (two patients), cardiac echogram (one patient), imaging of the abdominal cavity (four patients), 350
cerebrospinal fluid examination (two patients), imaging of the central nervous system (four patients), thyroid tests (nine patients), cortisolaemia (one patient), and detection of various circulating markers of cancer (one patient). Four patients received no treatment and ten received various combinations of anti-inflammatory drugs and antibiotics (table 1). Four patients were treated by different antibiotics, including co-trimoxazole (three patients), standard antituberculous antibiotic therapy (two patients), chloramphenicol (one patient), minocycline (one patient), azythromycin (one patient), and ofloxacin (one patient). Complete improvement with antibiotics alone was seen in one patient (patient 8, who received co-trimoxazole and antituberculous drugs pyrazinamide, ethambutol, isoniazid, and rifampicin). Two patients partially improved under antibiotic therapy (patient 1 received co-trimoxazole plus chloramphenicol; patient 7 received minocycline) and completely improved with subsequent administration of steroids. No improvement was seen in patient six with trimethoprim, azythromycin, or ofloxacin, each used alone. Non-steroidal anti-inflammatory drugs were ineffective. Steroids were used at various doses in eight of 14 patients and resulted in partial (three) or complete recovery (five) in a few days or weeks.
Discussion In several patients seen since 1993 with varied combinations of arthromyalgias, myopathic weakness, increased creatine kinase concentrations, and raised erythrocyte sedimentation rate, muscle biopsy had shown a unique pattern of conspicuous accumulations of large macrophages with fine PAS-positive granular cytoplasm in the epimysium, perimysium, or endomysium, associated with mild CD8 T-cell infiltrates, and minimum muscle fibre damage. PAS-positive macrophages were not detected elsewhere other than in muscle, fascia, and the deep subcutaneous tissue, although symptoms suggesting possible involvement of other organs were present in a few patients. Routine investigations did not show the cause of the disorder. Patients responded well to steroids, with or without antibiotics. Unlike sarcoid granulomatas, macrophages infiltrating muscle connective tissue did not have an epithelioid appearance, did not form giant cells, and were not organised in nodules. Macrophagic myofasciitis was, therefore, easily distinguishable from sarcoid myopathy,3 and was clearly different from the fasciitis-panniculitis syndromes, such as Shulman’s disease, toxic-oil syndrome, and eosinophilia-myalgia syndrome, in which macrophages epithelioid cells, and giant cells may accompany chronic inflammation, fibrosis, and vascular changes but never predominate in the infiltrate.10 With the exception of sarcoid-like granulomatous myopathies, conspicuous accumulations of histiocytes in skeletal muscle are extremely rare. Histiocytes may derive from Langerhans or dendritic cells, or, as in our patients, from monocytes or macrophages.11 Other than in sarcoidosis and sarcoid-like disorders12 “reactive” accumulations of macrophages occur in storage diseases related to inborn errors of metabolism,13 benign proliferations of macrophages known as non-Langerhans histiocytoses,11,14 and infectious diseases associated with a defect in macrophage handling of bacteria.15
THE LANCET • Vol 352 • August 1, 1998
ARTICLES
In our patients, the nature of the accumulated material in macrophages remained undetermined, and, therefore, a storage disease should be considered. However, the histories of our patients gave no support for an inherited metabolic disease, or for accumulation of an organic material of iatrogenic origin. Light-microscopy and electron microscopy characteristics of the accumulated material were not those of the ceroid pigment found in sea-blue histiocytes, and the disorder seemed to be an inflammatory disease rather than a metabolic disorder. We are aware of no storage disease that selectively involves macrophages and produces symptoms in muscles. Non-Langerhans histiocytoses may present with 11 different clinicopathological patterns.11 One of these patterns, multicentric reticulohistiocytosis, may be associated with ill-defined muscle involvement.16 Multicentric reticulohistiocytosis, however, cannot be confused with macrophagic myofasciitis since nodular skin lesions, severe destructive arthropathy, and abundant multinucleated giant-cell formation are the hallmarks of multicentric reticulohistiocytosis.11,14 Infectious diseases associated with macrophage dysfunction are generally characterised by the presence of PAS-positive material within macrophages, as in macrophagic myofasciitis. These diseases include malakoplakia,17 Whipple’s disease,18 and M avium complex infection.19,20 Malakoplakia is a rare chronic inflammatory disorder associated with chronic infections, mainly by coliform bacilli, characterised by collections of macrophages in the genitourinary tract and, less frequently, in other sites.17 This disorder is not thought to cause muscle disease.3 Histopathologically, intracellular and extracellular calcified bodies with a laminar structure, known as Michaelis-Gutmann bodies, are the pathognomonic feature of the disease.17 Such inclusions were not seen in our patients, and, therefore, macrophagic myofasciitis should not be diagnosed as muscle malakoplakia. We initially considered the diagnosis of Whipple’s disease for many of our patients because of the report of one case of progressive myopathy associated with this disease,21 and the rare occurrence of Whipple’s disease without intestinal involvement (estimated at ten of 760 patients18). In the reported patient with myopathy, muscle biopsy showed macrophages containing typical PAS-positive sickle-shaped particles in the perimysium.21 Intracytoplasmic membranous bodies and bacillary bodies were found on electronmicroscopy in macrophages in muscle and jejunum. Oral antibiotics improved the myopathy. We are aware of only one other reference to specific myopathy during Whipple’s disease.22 None of the seven myopathologists who were coinvestigators in our study has seen such a case. A clinical review of 52 French patients with Whipple’s disease did not mention muscular symptoms, even nonspecific symptoms.23 It is unlikely that our patients had Whipple’s disease since they had no digestive symptoms, no PAS-positive macrophages in the lamina propria of the digestive tract, of T whippelii in the jejunum or muscle, and no ultrastructural evidence of bacillary bodies within or in the vicinity of macrophage infiltrates in muscle. A few patients infected by non-tuberculous mycobacterial infections may develop a pronounced
THE LANCET • Vol 352 • August 1, 1998
systemic histiocytic proliferation that mainly affects lymphoid organs and the gastrointestinal tract.19,20 Specific muscle involvement by the M avium complex has rarely been observed in patients who have died from AIDS.24 The histological picture has not generally indicated severe necrosis or giant-cell reaction, but has been characterised by the presence of abundant intracellular acid-fast bacilli, easily detectable by special stains and electron microscopy.19,20,24 Ziehl-Neelsen and auramin stains as well as PCR detection of M avium complex were negative in our patients, thus precluding the diagnosis of M avium complex infection. Finally, macrophagic myofasciitis does not correspond to any of the previously described histiocytoses or reactive macrophage accumulations. At present the cause of this unique disorder is unknown. Most emerging or re-emerging diseases, however, are or will probably be caused by infectious agents.1 The spectrum of recognised mycobacterial diseases, has, for example, increased in the past few years.25 Indeed, there are some indirect arguments for an infectious cause of macrophagic myofasciitis. Exposure to tuberculosis was noted in two patients, of whom one improved completely under antituberculous quadritherapy combined with cotrimoxazole. Like patients with malakoplakia, Whipple’s disease, or histiocytosis associated with M avium complex, some of our patients had an altered immunological status. Four patients had concomitant systemic disorders—malignant tumour in one, and autoimmune disease in three. One patient had monocytosis, and two had a persistent lymphocytopenia consistent with an altered cell-mediated immunity. A striking feature was that five patients took chloroquine or hydroxychloroquine. Impaired antimycobacterial activity of macrophages is linked to impaired secretion of tumour necrosis factor ␣ and interleukins 1-␣ and 1-, 6, and 12 by macrophages and impaired secretion of interferon-␥ by T lymphocytes.15,26–28 Interestingly, hydroxychloroquine inhibits secretion of interleukins 1-␣ and 6 by macrophages.29 This action may interfere with the complex cytokine interplay regulating antimycobacterial activity and may be harmful; similar effects may be seen with other drugs that affect host defence against mycobacteria.27 Further work is needed to identify the cause of macrophagic myofasciitis, but the disorder is potentially treatable. Contributors R K Gherardi, L Belec, and M Fardeau designed the study. P Cherin and F-J Authier were responsible for clinical investigations. R K Gherardi, M Coquet, F-J Authier, D Figarella-Branger, J-M Mussini, J-F Pellissier, and M Fardeau were responsible for histopathological invstigations. L Belec did molecular microbiological assessments. R K Gherardi and F-J Authier wrote the paper. All authors read and edited the manuscript.
Acknowledgments We thank D Bernard, J M Berthelot, C Bihel-Schleich, R Carlioz, J-Y Colin, B Eymard, F Gray, P Heine, S Herson, J M Labrousse, J-P Leroy, X Pennaforte, D Ranoux, D Sereni, A Schaeffer, J A Urtizberea, F Vaylet, and P-J Weller, who referred or contributed to the assessment of the patients. This work was supported by the Association Française contre les Myopathies (AFM).
References 1
2
Schwartz DA. Emerging and reemerging infections: progress and challenges in the subspecialty of infectious disease pathology. Arch Pathol Lab Med 1997; 121: 776–84. Dalakas MC. Polymyositis, dermatomyositis and inclusion body myositis. N Engl J Med 1991; 325: 1487–98.
351
ARTICLES 3
4
5
6
7 8
9
10
11 12
13 14
15
16
17
Banker BQ. Other inflammatory myopathies. In: Engel AG, Franzini-Armstrong, eds. Myology, 2nd edn, vol 2. New York: McGraw-Hill, 1993: 1461–86. Alonzo-Ruiz A, Zea-Mendoza AC, Salazar-Vallinas JM, Rocamora-Ripoll A, Beltran-Guttierrez D. Toxic oil syndrome: a syndrome with features overlapping those of various forms of scleroderma. Semin Arthritis Rheum 1986; 15: 200–12. Kaufman LD, Seidman RJ, Gruber BL. L-tryptophan-associated eosinophilic perimyositis, neuritis and fasciitis: a clinicopathogic and laboratory study of 25 patients. Medicine 1990; 69: 187–99. Mayeno AN, Benson LM, Naylor S, Colberg-Beers M, Puchalski JT, Gleich GJ. Biotransformation of 3-(phenylamino)1,2-propanediol to 3-(phenylamino)alanine: a chemical link between toxic oil syndrome and eosinophilia-myalgia syndrome. Chem Res Toxicol 1995; 8: 911–16. Gherardi R. Skeletal muscle involvement in HIV-infected patients. Neuropathol Appl Neurobiol 1994; 20: 232–37. Dalakas MC, Illa I, Pezeshkpour GH, Laukaitis JP, Cohen B, Griffin JL. Mitochondrial myopathy caused by long-term zidovudine therapy. N Engl J Med 1990; 322: 1098–05. Mhiri C, Baudrimont M, Bonne G, et al. Zidovudine myopathy: a distinctive disorder associated with mitochondrial dysfunction. Ann Neurol 1991; 29: 606–14. Naschitz JE, Boss JH, Misselevitch I, Yeshurun D, Rosner I. The fasciitis-panniculitis syndromes: clinical and pathologic features. Medicine 1996; 75: 6–16. Snow JL, Daniel Su WP. Histiocytic disorders. J Am Acad Dermatol 1995; 33: 111–16. Friedlander JS, Weatherall DJ, Ledingham JGG. A chronic granulomatous syndrome of unknown origin. Medicine 1990; 69: 325–31. Cline MJ. Histiocytes and histiocytosis. Blood 1994; 84: 2840–53. Zellger BWH, Sidoroff A, Orchard G, Cerio R. Non-Langerhans cells histiocytosis: a new unifying concept. Am J Dermatopathol 1996; 18: 490–504. Marth T, Neurath M, Cuccherini BA, Strober W. Defects of monocyte interleukin 12 production and humoral immunity in Whipple’s disease. Gastroenterology 1997; 113: 442–48. Gao IK, Goronzy JJ, Weyand CM. Ten year follow-up of a patient with multicentric reticulohistiocytosis associated with myopathy. Scand J Rheumatol 1990; 19: 437–41. Lack EE. Malakoplakia. In: Connor DH, Chandler FW, Schwartz DA, Manz HJ, Lack EE, eds. Pathology of infectious
352
diseases, vol 2. Stamford, USA: Appleton and Lange, 1997: 1647–54. 18 Dobbins WO III. Whipple’s disease. In: Connor DH, Chandler FW, Schwartz DA, Manz HJ, Lack EE, eds. Pathology of infectious diseases, vol 1. Stamford, USA: Appleton and Lange, 1997: 897–903. 19 Miranda D, Vuletin JC, Kauffman SL. Disseminated histiocytosis and intestinal malakoplakia: occurrence due to Mycobacterium intracellulare infection. Arch Pathol Lab Med 1979; 103: 302–05. 20 Gillin JS, Urmacher C, West R, Shike M. Disseminated Mycobacterium avium-intracellulare infection in acquired immunodeficiency syndrome mimicking Whipple’s disease. Gastroenterology 1983; 85: 1187–91. 21 Swash M, Schwartz MS, Vandenburg MJ, Pollock DJ. Myopathy in Whipple’s disease. Gut 1977; 18: 800. 22 Banker BQ, Engel AG. Basic reactions of muscle. In: Engel AG, Franzini-Armstrong, eds. Myology, 2nd edn, vol 1. New York: McGraw-Hill, 1993: 832–88. 23 Vital Durand D, Lecomte C, Cathébras P, Rousset H, Godeau P, SNFMI research group on Whipple’s disease. Whipple disease: clinical review of 52 cases. Medicine 1997; 76: 170–84. 24 Wrzolek MA, Sher JH, Kozlovski PB, Rao C. Skeletal muscle pathology in AIDS: an autopsy study. Muscle Nerve 1990; 13: 508–15. 25 Watt B. Lesser known mycobacteria. J Clin Pathol 1995; 48: 701–05. 26 Michelini-Norris MB, Blanchard DK, Pearson CA, Djeu JY. Differential release of interleukin (IL)-1 beta, and IL-6 from normal human monocytes stimulated with a virulent and an avitulent isogenic variant of Mycobacterium avium-intracellulare complex. J Infect Dis 1992; 165: 702–09. 27 Sathe SS, Sarai A, Tsigler D, Nedunchezian D. Pentoxifylline aggravates impairment in tumor necrosis factor-alpha secretion and increases mycobacterial load in macrophages from AIDS patients with disseminated Mycobacterioum avium-intracellulare complex infection. J Infect Dis 1994; 170: 484–87. 28 Frucht DM, Holland SM. Defective monocyte costimulation for IFN-␥ production in familial disseminated Mycobacterium avium complex infection: abnormal IL-12 regulation. J Immunol 1996; 157: 41–16. 29 Sperber K, Quraishi H, Kalb TH, Panja A, Stecher V, Mayer L. Selective regulation of cytokine secretion by hydroxychloroquine: inhibition of interleukin 1 alpha (IL-1 alpha) and IL-6 in human monocytes and T-cells. J Rheumatol 1993; 2): 803–08.
THE LANCET • Vol 352 • August 1, 1998