Clinical Radiology (1990)41, 53-56

Magnetic Resonance Differentiation of Acute and Chronic Osteomyelitis in Children M. D. COHEN, D. A. CORY, M. K L E I M A N , * J. A. S M I T H and N. J. B R O D E R I C K

Departments of Radiology and *Pediatrics, Riley Hospital for Children, 702 Barnhill Drive, Indianapolis, I N 46202, USA

A comparison has been made of the M R I findings of 17 studies in 16 patients with osteomyelitis: eight studies were performed in patients with acute osteomyelitis, six in chronic and three in healed osteomyelitis. Soft tissues, cortex and marrow were assessed in all patients. The best predictors of acute osteomyelitis were poorly defined soft tissue planes, absence of cortical thickening, and a poor interface between normal and diseased marrow. In contrast, chronic osteomyelitis was suggested by the presence of welldefined soft tissue abnormality, thickened cortex, and a relatively good interface between normal and diseased marrow. The appearance of osteomyelitis did not vary in different anatomic sites.

Osteomyelitis is a potentially debilitating infection which, if diagnosed early, can be cured with antibiotic therapy and, sometimes, operative drainage and debridement. Complications of delayed or inadequate treatment include bone deformity, avascular necrosis of the femoral head and chronic drainage. The diagnosis of osteomyelitis, though frequently straightforward, can still be delayed (MaAfee and Samin, 1985; Modic et al., 1986; Unger et al., 1988). Radiographs are usually normal in the early stages of the disease. When abnormal, the findings can be nonspecific (Modic et al., 1986; Beltran et al., 1987). Radionuclide studies are reasonably sensitive, but miss approximately 10% of cases of acute osteomyelitis; positive findings are non-specific (MaAfee and Samin, 1985; Beltran et al., 1987; Unger et al., 1988). Magnetic resonance (MR) has been utilised to evaluate osteomyelitis (Fletcher et al., 1984; MaAfee and Samin, 1985; Modic et al., 1986; Beltran et al., 1987; Ehman et aI., 1988; Unger et al., 1988), but there are few reports of the features that allow differentiation of acute and chronic osteomyelitis in children. The objective of this study was to evaluate and describe the magnetic resonance findiflgs in the soft tissues, bone cortex, and bone marrow in patients with acute, chronic or healed osteomyelitis.

firmed. In 13 patients there was histologic and/0r microbiologic confirmation of osteomyelitis, and in three there was a strongly compatible clinical course with an appropriate response to antibiotics. Two of the latter had other imaging studies which were compatible with the diagnosis of osteomyelitis. Bone marrow disease was diagnosed if there was a decrease in signal intensity visualised on T1 images. Cortical thickening was diagnosed by comparison with the opposite side. Chronic infection was diagnosed by a continuation of clinical symptoms over 3 months in duration, the presence of bony sclerosis and thickening on plain film radiograph, and/or pathological findings of chronic inflammation or sequestrum. Seventeen studies were performed in 16 patients; one patient had two bones affected. The age range was 5 months to 15 years. Infections were of the femur (5), calcaneus (4), tibia (2), humerus (2), ischial tuberosity (1), ulna (1), a metatarsal (1), and glenoid (1). Eight patients had acute osteomyelitis, two of whom had received antibiotics at the time of M R imaging. Six patients had chronic osteomyelitis. Three patients were studied in the healed phase, one of whom had also been studied during the acute phase.

RESULTS There were no differences in the appearance of the osteomyelitis based on anatomic location. Soft Tissues

Soft tissue abnormalities were identified in all patients (Figs 1-4) except two with healed disease. The third

METHODS Seventeen M R studies in 16 patients with osteomyelitis were reviewed. Seven studies were performed with a Technicare 0.15 Tesla unit, and 10 with a Picker 1.5 Tesla unit. At least two imaging planes were used. T1 weighted and T2 weighted images were obtained in 15 patients; the other patient had only T1 images. The study included all patients, seen over a 5 year period, who had M R scans, and in whom the diagnosis of osteomyelitis was conCorrespondence to: Mervyn D. Cohen, Department of Radiology, RileyHospitalfor Children, 702 BarnhillDrive,Indianapolis,IN 46202, USA.

Fig. 1 - Acute osteomyelitis.T2 image (1000/60) shows extensivesoft tissue swellingaround the distal humerus. The margins of the soft tissue swelling are relativelypoorly defined,particulary laterally.

54 Table 1 - M R spectrum of

CLINICAL RADIOLOGY

osteomyelitis Stage of osteomyelitis

MR findings

Acute (n = 8)

Chronic (n = 6)

Soft tissues Margins of abnormality: Well defined (n = 5) Poorly defined (n = 9)

1 7

4 2

Extent: + ++ +++

1 4 3

3 2 I

Cortex Normal Periosteal reaction Thickened Sequestrum

8* 0 0 0

0 0 4 2

Junction with normal marrow Well defined (n = 4) Poorly defined (n = 5) Not evaluated (n = 5)

1 3 4

3 2 1

(a)

* One of these patients had a mild periosteaI reaction on a radiograph.

patient with healed femoral head osteomyelitis had presented with new hip pain due to aseptic necrosis of the femoral head. The soft tissue abnormality in this patient was attributed to this complication although pathological proof was not available. Margins of the soft tissue abnormalities were well defined in 4 of 6 patients with chronic (Fig. 3) and l of 7 patients with acute (Figs '1 & 2) osteomyelitis (Table 1). The extent of the soft tissue abnormality varied from a thin rim adjacent to the infected bone (Fig. 3) to a large extensive abnormality (Figs 1 & 2). Soft tissue involvement tended to be less in patients with chronic disease (Table 1). Bone Cortex

A mild periosteal reaction evident on plain radiographs in 1 of the 8 patients with acute disease (Table 1) could not be identified on M R images. Cortical thickening in four patients with chronic disease (Table 1) was clearly identified on M R as low signal intensity on all pulse sequences. Sequestra, present in two patients, could be identified on M R images, but less readily than with CT or plain film radiography.

(b)

(c)

Fig. 2 Acute osteomyelitis of calcaneus. (a) Plain film radiograph reveals no definite abnormality in the calcaneus. The soft tissues are swollen. (b) T 1 (500/30) weighted image shows patchy loss of signal from almost all of the bone marrow in the calcaneus. There are no margins between normal and abnormal marrow. The soft tissues are swollen and of decreased intensity compared to normal. The interface between muscles and subcutaneous fat is lost and there is patchy decreased signal fn the fat due to inflammatory oedema. This image was obtained on the same day as image A. (c) The opposite normal foot is shown for

comparison.

Marrow

Marrow abnormalities were identified in all patients with acute or chronic osteomyelitis. Marrow abnormalities were most readily identified on T1 weighted images (Figs 2, 3 & 4). The abnormality consisted of moderate to marked decrease of signal intensity compared to a strong signal in normal marrow. The interface between diseased and normal marrow was evaluated (Table 1). In five studies this assessment was not possible either because of inappropriate imaging planes or because the entire marrow o f the affected bone was abnormal. The interface between normal and abnormal marrow was evaluated in 9 patients with acute or chronic osteomyelitis (Table 1, Figs 2, 3 & 4). One of four patients with acute and 3 of 5

patients with chronic disease had a well defined interface (Table 1). Three images were obtained in patients with healed osteomyelitis. In two of these patients the M R image was completely normal. The third Patient had soft tissue and marrow abnormalities, which were attributed to avascular necrosis of the femoral head complicating earlier healed osteomyelitis. DISCUSSION Magnetic resonance can demonstrate abnormality in osteomyelitis (Fletcher et al., 1984; MaAfee and Samin,

MR DIFFERENTIATIONOF ACUTE AND CHRONIC OSTEOMYELITIS

55

(a) Fig. 3 - Chronic osteomyelitis of the tibia. (a) Plain film radiograph shows cortical thickening over the anterior aspect of the right tibia. A focal area of decreased signal intensity is also identified. (b) Transverse T1 weighted image shows reduction of signal from the marrow of the tibia at the level Of the lesion seen on the plain film. Cortical thickening is also identified as low signal region anteriorly. (e) Coronal T1 image (700/26) shows more extensive marrow abnormality than would be expected from the plain film.The marrow abnormality extends to about 2 cm from the proximal growth plate. At this level there is a fairly sharp demarcation between low intensity, inflamed marrow and adjacent normal marrow. (d) Transverse T2 image shows increased signal in the bone marrow as compared to the TI image. Note the thin rim of increased signal in the soft tissues. This is not apparent on the T 1 image.

(a)

(b)

(c)

1985; M o d i c et al., 1986; Beltran et al., 1987; E h m a n et al., 1988; U n g e r et al., 1988). T h e different appearance of acute a n d chronic disease has however n o t been described in detail. I n the present study M R was a b n o r m a l in all patients with acute or chronic osteomyelitis. C h r o n i c osteomyelitis is suggested by a good interface between n o r m a l a n d a b n o r m a l b o n e m a r r o w (60%), cortical thickening_+ sequestrum (100%) a n d well defined b o u n daries to the soft tissue a b n o r m a l i t y (66%). A p o o r interface between n o r m a l a n d a b n o r m a l m a r r o w (75%), absent cortical thickening (100%), a n d poorly defined soft tissue a b n o r m a l i t y (87.5%), should suggest acute disease. The extent of soft tissue a b n o r m a l i t y tends to be less in those patients with chronic disease. There is, however, overlap between the a p p e a r a n c e of acute a n d chronic osteomyelitis. Healed osteomyelitis appeared n o r m a l o n M R unless there was a c o m p l i c a t i o n such as ava'scular necrosis.

Fig. 4 - (a)

56

CLINICAL RADIOLOGY

The present study suggests that there are differences on M R between acute, chronic and healed osteomyelitis and indicates that further studies in a larger population group would be valuable.

REFERENCES

(b)

¢) Fig. 4 - Partially treated chronic osteomyelitis. (a) Radiograph shows decreased density in the posterior/inferior aspect of the calcaneus. (b) T 1 weighted MR image shows extensive low signal affecting about half of the posterior/inferior aspect of the calcaneus. The lesion is more extensive than that seen on the plain film radiograph. The margins between normal and diseased bone marrow are moderately well defined. There is mild swelling and patchy irregularity of signal intensity in the overlying fat. (e) T2 weighted image (2000/120) shows increased signal in the affected area of the calcaneus. The extent of disease is even more than was evident on the T1 weighted image.

Beltran, J, Noto, A, McGhee, R, Freedy, RM & McCalta, MS (1987). Infections of the musculoskeletal system: high-field-strength MR imaging. Radiology, 164, 449-454. Ehman, RL, Berquist, T & McLeod, R (1988). MR imaging of the musculoskeletal system: a 5-year appraisal. Radiology, 166, 313-320. Fletcher, BD, Scoles0 PV & Nelson, AD (1984). Osteomyelitis in children: detection by magnetic resonance. Radiology, 150, 57 60. MaAfee, JG & Samin, A (1985). In-111 labelled leukocytes: a review of problems in image interpretation. Radiology, 155, 221-229. Modic, MT, Pflanze, W, Feiglin, D & Belhobek, G (1986). Magnetic resonance imaging of musculoskeletal infections. Magnetic Resonance Imaging of the Museuloskeletal System, 24, 247 258. • Unger, E, Moldofsky, P, Gatenby, R. (1988). Diagnosis of osteomyelitis by MR imaging. American Journal of Roentgenology, 150, 605 610.

Magnetic resonance differentiation of acute and chronic osteomyelitis in children.

A comparison has been made of the MRI findings of 17 studies in 16 patients with osteomyelitis: eight studies were performed in patients with acute os...
3MB Sizes 0 Downloads 0 Views