A previously healthy twenty-five-year-old woman presented to our emergency department with incapacitating and worsening backache, without radiation of pain to the lower extremities, for three days. She denied any recent febrile illness, loss of weight, or loss of appetite. There was no history of trauma or infection, local injection, or foreign body.
The patient was afebrile and the vital signs were normal. Neurological examination showed grade 5/5 motor strength in the lower extremities with no sensory impairment. All extremities had normal deep tendon reflexes. The Babinski sign was absent bilaterally, and there was no ankle clonus. Findings on the straight leg raise test were negative. Local examination of the back revealed notable paraspinal tenderness on the left side, extending from the lower thoracic spine down to the midlumbar spine, with no redness, fluctuation, or sinuses. No other swellings were noted on the body. Pulmonary and cardiovascular examinations revealed no abnormalities, and bowel and bladder function were normal. On laboratory workup, the white blood-cell count was normal. The erythrocyte sedimentation rate was 80 mm/h (normal, 0 to 20 mm/h), and the C-reactive protein level was 3.05 mg/dL (normal, 0 to 0.5 mg/dL). The creatine phosphokinase level measured 56 U/L (normal, 20 to 120 U/L). The CT scan of the lumbar spine yielded no evidence of a substantial abnormality or soft-tissue swelling. The patient was admitted to the orthopaedics department for additional evaluation and pain management.
Despite treatment with intravenous morphine, the patient could barely walk without assistance. MRI scans of the lumbar spine (T2-weighted and short tau inversion recovery [STIR]) revealed increased signal intensity of the paraspinal muscles of the L3-L5 vertebrae on the left and a small effusion in the left L3-L4 facet joint (Figs. 1 and 2). No intramuscular abscesses were found, and the signal in the paraspinal muscles on the contralateral side was normal. There were no abnormalities on abdominal ultrasound, and no enlargement of the retroperitoneal lymph nodes was evident. On repeated laboratory tests, the white blood-cell count was within normal range, but the C-reactive protein level and erythrocyte sedimentation rate remained elevated.
A clinical diagnosis of infection involving the paraspinal muscles and possibly the facet joint was suspected. Since the MRI scan revealed only subtle changes, which could possibly be explained by the prolonged supine position of the patient, she was referred for a PET-CT with 18-FDG (Fig. 3). FDG uptake was increased in the left longissimus muscle near the L3-L4 vertebrae, with involvement of the L3-L4 left facet joint, as shown on MRI. The preliminary diagnosis was infection of the paraspinal muscles. A CT-guided muscle biopsy was performed on day eight of hospitalization, and the tissues were submitted for pathologic study and cultures. The cultures were negative; the patient had not been treated with antibiotics prior to the biopsy and cultures. Pathologic findings in the involved longissimus muscle included edema and inflammation of the perimysium, with abundant lymphocytes and scattered eosinophils. Immunohistochemical stains for cluster of differentiation 3 (CD3), CD4, CD8, CD20, and CD68 revealed a predominant population of T cells within the infiltrate, mainly CD4 cells, with scattered B cells and histiocytes (Figs. 4 and 5). Soon after the biopsy sample was taken, the patient was started on empiric treatment with intravenous cloxacillin and oral ciprofloxacin for the most likely causative bacteria (Staphylococcus aureus)5. After two days, she reported improvement in the backache and was able to walk without assistance. After seventy-two hours, the C-reactive protein level declined. Blood and tissue cultures were negative. The patient remained afebrile throughout hospitalization.
The patient was discharged and began home treatment with intravenous cefazolin and oral ciprofloxacin for four weeks. She was followed regularly in the outpatient clinic for one year. The condition resolved completely; an MRI scan performed at the last follow-up visit revealed no abnormalities.
Because of the rarity of pyomyositis in temperate climates, the common lack of specific signs or symptoms, and the frequent negative blood cultures, a considerable delay in the diagnosis of pyomyositis is not uncommon10. The patient described in this case report presented early after the onset of back pain, with a clinical and radiographic picture typical of the invasive stage of pyomyositis. However, the only signs that raised suspicion of an infection at the time were an elevated C-reactive protein level and a high erythrocyte sedimentation rate. Blood culture results and CT-guided biopsy did not reveal any pathogen, although they showed a mild inflammatory process. This is not surprising given that causative bacteria are frequently not isolated from blood cultures or cultures of purulent material in primary pyomyositis5. The source of infection in the majority of patients described in the literature, as well as in the patient presented here, is unknown. It is believed to be a complication of transient bacteremia without an obvious penetrating injury or other portal of entry5.
Myositis can be an infectious, noninfectious, or systemic process. Noninfectious, nonsystemic myositis conditions include dermatomyositis, polymyositis, or inclusion body myositis11-13. All three were ruled out in our patient by the findings of solitary involvement of the unilateral paraspinal muscles, the involvement of the perimysium without substantial endomysial infiltration, and normal levels of creatine phosphokinase12,13.
A variety of imaging modalities can be used for the diagnosis of spinal infections. MRI is considered the modality of choice for patients with suspected pyomyositis, not only because of its sensitivity for detecting the disease but also for its utility in diagnosing osteomyelitis and abscess formation, which when present, may alter treatment14. However, if the MRI findings are equivocal, additional tests are needed. The use of 18-FDG PET-CT is a promising combined modality for the diagnosis of soft-tissue infections15. It provides functional and anatomical data simultaneously, improving the interpretation of the findings. However, we found only one case report in the literature advocating the use of PET-CT for the diagnosis of pyomyositis involving the calf muscles7, and no descriptions of its use for paraspinal pyomyositis. We used both MRI and 18-FDG PET-CT with our patient, which showed that the adjacent L3-L4 facet joint also was involved. This could explain the severity of the patient’s pain at presentation. The biopsy study revealed a mild inflammatory process without abundant neutrophils. This finding, which represents inflammation in the periphery of the lesion, may have been due to a sampling error.
Our patient had an acute infection of a single facet joint and the surrounding paravertebral muscles. She had no history of spinal pathology or immunosuppression, and no source of the infection was isolated. A thorough history combined with a meticulous physical examination, imaging studies, and tissue biopsy was necessary to establish the diagnosis. The suspected diagnosis was strongly supported by the PET-CT findings. Because of the rarity of this entity and the lack of specific clinical findings, PET-CT can be an additional, valuable tool that may assist in supporting the suspected diagnosis of paraspinal pyomyositis. Failure to recognize pyomyositis can lead to delayed and inappropriate management. In our patient, the outcome was favorable despite the severity of the muscle involvement.