Primary osteomyelitis of' the sternum is a rare condition. Infection may follow chest trauma and is a recognised complication of open heart surgery, with a reported incidence of 1-2%.' Since the detailed review by Wilensky and Samuels in 19262 there have' been only occasional reports. We describe a case of chronic sternal infection without preceding trauma and present a review of published reports. Case reportA 40 year old housewife presented with pain and swelling in the front of her chest five months after an acute self limiting diarrhoeal illness in Tenerife. The swelling had developed over five days and was associated with weight loss of 6.3 kg. Examination revealed an inflammed tender area over the manubrium measuring 3 x 2 cm and pyrexia of 37.40C. There was no lymphadenopathy and no other focus of infection was found. Radiographs showed widening of the sternum without lysis. The white cell count was 11.3 x 109/l; the initial erythrocyte sedimentation rate was 65 Westergren units in one hour, rising to 115 over the next week. Needle aspiration of the sternum showed scanty Gram negative micrococci, considered to be contaminants. Blood, stool, urine, and marrow cultures were negative and serological examination for brucella and typhoid organisms also gave negative results. Culture for acid fast bacilli was negative. A needle biopsy of the sternum was performed and histological examination showed features of chronic non-specific inflammation with reactive woven bone formation, but no organisms were seen. A gallium bone scan (figure), which has been of diagnostic value in previous cases,3 showed increased uptake along the sternum suggestive of osteomyelitis and treatment was started with fusidicacid on the assumption that Staphlococcus aureus was the most likely pathogen. There was no improvement in her condition after one month so treatment was discontinued and surgical exploration of the sternum was undertaken. Material obtained showed microscopic features of chronic inflammation. The curettings obtained did not contain pus and again no organisms were identified. Doxycycline was prescribed. Symptomatic relief was gradually achieved, the signs resolved, and the patient remains well several months later.

Objectives: To assess the physiological effect of lowpressure graded compression stockings (GCS) on the blood flow of the lower limb with the use of colour Doppler.Design: A randomised controlled study. Setting: X-ray Department in a teaching hospital. Patients: Forty-five healthy, adult, voluntary subjects. ethod: Subjects were randomised into three groups to either wear thigh-or knee-length GCS of the same type or no stockings. The diameter and cross-sectional area of the femoral and popliteal veins and the peak venous velocities were measured with colour Doppler before the application of GCS and after 20 min bed rest with the stockings in situ.Results: Measurements showed a significant increase of the peak velocity in the femoral vein with thigh-length GCS (p < 0.002). There was a significant decrease of the popliteal vein diameter and cross-sectional area with thigh-length GCS (p < 0.05). There was no significant change in the knee-length GCS group and the control group, in which patients did not wear stockings. A paired t-test was used to determine the significance of the changes.onclusions: Thigh-length GCS have an effect of Increasing venous peak velocity in the femoral vein, hence decreasing venous stasis. Furthermore, they decrease the dilatation of the popliteal vein, which may reduce the risk of intimal tears occurring, which contribute to venous thrombogenesis. Other methods can given more information in assessing the effect of various lengths of stocking on venous stasis.

No abstract