Article Text


Myositis associated graft-versus-host-disease presenting as respiratory muscle weakness
  1. A L Stephenson,
  2. I R A Mackenzie,
  3. R D Levy,
  4. J Road
  1. Department of Respirology, Department of Pathology, Vancouver General Hospital & Health Sciences Centre, The University of British Columbia, 2775 Heather Street, Vancouver, BC, V5Z 3J5 Canada
  1. Dr J Road, Department of Respirology, Vancouver General Hospital, 2775 Heather Street, Vancouver, British Columbia V52 3J5, Canada


Myositis associated with graft-versus-host-disease (GVHD) typically presents with proximal muscle weakness, myalgias, and a raised creatinine phosphokinase (CPK) level. We report a case of a 51 year old man who developed respiratory muscle weakness five years after an allogeneic bone marrow transplant for multiple myeloma. His symptoms included tachypnoea, abdominal paradox, and orthopnoea. Pulmonary function tests revealed diminished vital capacity and maximal inspiratory and expiratory pressures. Serum CPK levels were raised and a peripheral muscle biopsy specimen was consistent with GVHD. He improved with immunosuppressive therapy.

  • graft-versus-host-disease
  • myositis
  • respiratory muscle weakness

Statistics from

Graft-versus-host disease (GVHD) is a major cause of morbidity and mortality following allogeneic bone marrow transplantation (BMT); 20–50% of BMT recipients who survive more than 100 days after BMT will develop chronic GVHD.1 Although myositis has been described in association with GVHD, it is rare.2 It may be the sole manifestation of active GVHD although patients typically have other organ involvement in addition to muscle.3 Myositis associated with GVHD presents in a fashion similar to idiopathic myositis with proximal muscle weakness, myalgias or muscle pain, and an increased creatinine phosphokinase (CPK) level.4 To our knowledge, there is only one previously published case report of GVHD myositis involving the respiratory muscles.5 This differed from our case in that GVHD developed rapidly two months after a donor leucocyte transfusion (DLT) for relapse of acute myeloblastic leukaemia and the DLT was considered a possible cofactor in the myositis. The histopathological features of GVHD myositis typically show degenerating muscle fibres and necrosis with a perimysial monocytic/lymphocytic infiltrate with a predominance of CD8+ cytotoxic cells.4 Response to treatment with prednisone, with or without cyclosporin, is dramatic with marked improvement of weakness within days of initiating treatment.

Case report

A 51 year old man who was a non-smoker was admitted to the hospital with a three month history of progressive orthopnoea and cough. He had undergone an allogeneic BMT from a matched sibling donor for multiple myeloma in 1993. His course was complicated by GVHD involving the skin, bronchiolitis obliterans of the lungs, pulmonary embolism, and central nervous system toxoplasmosis. On admission his medications included prednisone 20 mg alternating with 15 mg daily, trimethoprim-sulphamethoxazole 20 ml twice daily Saturday and Sunday, clindamycin 300 mg three times daily three times per week, pyrimethamine 25 mg three times per week, leucovorin 5 mg daily, fluticasone 250 μg twice daily, and salbutamol four times daily. Clinically he was tachypnoeic, using accessory muscles of respiration as well as having abdominal paradox, particularly in the supine position. He was noted to have slightly decreased peripheral muscle strength in the upper extremities. The deltoid, triceps, and biceps muscles were graded 4/5 while the muscle groups in the lower limbs were normal strength. A chest radiograph did not show evidence of active pulmonary disease. Arterial blood gas tensions showed evidence of hypoventilation with pH 7.39, Paco 2 6.4 kPa, Pao 2 10.5 kPa, and HCO3 29 mmol/l on room air. Nocturnal nasal BIPAP was initiated in hospital with symptomatic improvement of his orthopnoea.

Blood tests showed markedly raised CPK levels and pulmonary function tests revealed significantly decreased vital capacity and severely reduced maximum inspiratory and expiratory pressures (table 1) consistent with respiratory muscle weakness. The vital capacity was performed in the upright position and hence probably overestimated the supine vital capacity by at least 30%.6 Chest fluoroscopy showed severely diminished (virtually absent) bilateral diaphragm motion. No increase in diaphragmatic excursion was seen during the sniff test. A computed tomographic scan of the neck and mediastinum did not show any lesion along the expected course of the phrenic nerve to explain his diaphragmatic weakness. Electromyography (EMG) showed positive sharp waves and fibrillation in the left biceps, triceps, infraspinatus, right biceps, diaphragm, and the paraspinal muscles consistent with either myositis, axonal degeneration, or neuromuscular junction dysfunction. A muscle biopsy of the left triceps showed evidence of acute and chronic inflammation with associated muscle fibre necrosis (fig 1) The actively necrotic areas were infiltrated by both polymorphic leucocytes and mononuclear cells. Gram stain and culture of the tissue failed to reveal any infectious aetiology.

Table 1

Sequential values for blood tests and pulmonary function tests

Figure 1

Muscle biopsy specimen of left triceps showing evidence of acute and chronic inflammation associated with muscle necrosis.

The results of the investigations were interpreted as indicating an inflammatory myositis involving not only the proximal muscles, but both the inspiratory and expiratory respiratory muscles, particularly the diaphragm. His prednisone was increased to 60 mg daily and one day later he was started on cyclosporin 50 mg twice daily and mycophenolate mofetil (MMF) 1 g twice a day. His symptoms improved dramatically over the following week and his CPK level normalised. Repeat pulmonary function testing showed improved muscle strength with increased vital capacity and normal inspiratory and expiratory pressures. Lung volumes measured on 27 August (table 1) revealed total lung capacity of 5.51 l (77% predicted), residual volume of 2.47 l (117% predicted), and normal transfer factor. One month after starting treatment he no longer needed nocturnal BIPAP and his orthopnoea had completely resolved.


Inflammatory myopathy has been seen with chronic GVHD, although involvement of the respiratory muscles is rare with only one case report found in the literature. Five years after BMT our patient developed orthopnoea, general muscle weakness, and increased CPK levels with evidence on clinical examination, pulmonary function testing, and EMG studies of respiratory muscle weakness, particularly of the diaphragm. Although the EMG results are non-specific, they can be seen in inflammatory myopathies caused by GVHD. He responded dramatically to prednisone, cyclosporin, and MMF with resolution of his symptoms, normalisation of his CPK, and improvement of his vital capacity and respiratory muscle strength to the normal range. Interestingly, the spirometric parameters, as well as the inspiratory and expiratory pressures, improved between 22 July and 11 August before his immunosuppression was increased. His corticosteroid inhaler dose was increased during that period as well as his use of β agonists, which may have contributed to an improvement in his forced expiratory volume in one second (FEV1) and forced vital capacity (FVC). Respiratory muscle strength may have improved following the institution of nocturnal BIPAP which began on 24 July. BIPAP is known to decrease the work of breathing which may have allowed the respiratory muscles to recover partially and hence improve their function.

The muscle biopsy specimen showed evidence of muscle necrosis and an inflammatory infiltrate. The inflammatory cells seen were both neutrophils and monocytes which is somewhat atypical of GVHD myositis. Reported cases have documented a lymphocytic/monocytic predominant infiltrate on the muscle biopsy specimen. The presence of neutrophils in this case raised the possibility of an infectious process although no organism was cultured. Toxoplasma infection is a recognised risk in the immunocompromised host. Increased toxoplasma IgM levels suggestive of an active infection have been reported in idiopathic polymyositis but not in GVHD myositis.7 Our patient had a history of CNS toxoplasmosis and was on suppressive therapy with clindamycin, pyrimethamine, and leucovorin. No toxoplasma organism was cultured in blood, cerebrospinal fluid, or muscle biopsy, making active infection very unlikely in this patient. In addition, increasing the steroid dose would have worsened infection if present, but instead his clinical picture improved. Other possible causes for his weakness include steroid myopathy, but this seems unlikely given his clinical response to augmented steroids.

In conclusion, myositis associated with GVHD typically affects the proximal muscles, but it can also affect the respiratory muscles and cause significant morbidity in this population. The one published case report of GVHD involving the respiratory muscles resulted in respiratory failure and death soon after a donor leucocyte transfusion.5 It is possible that the donor T cells were responsible for causing severe GVHD. Our case illustrates that BMT patients are at risk for respiratory muscle involvement with GVHD and that this can be an insidious process occurring many years after the BMT. Furthermore, once muscle atrophy occurs, it has been suggested that the muscle may not recover despite increased immunosuppression and control of the inflammatory process. For this reason, a high clinical index of suspicion is needed for accurate diagnosis so that appropriate treatment can be instituted promptly.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.