980 THE JOURNAL OF BONE AND JOINT SURGERY  ONCOLOGY The final diagnosis in patients with a suspected primary malignancy of bone A. M. Malhas, R. J. Grimer, A. Abudu, S. R. Carter, R. M. Tillman, L. Jeys From The Royal Orthopaedic Hospital, Birmingham, United Kingdom  A. M. Malhas, MRCS, MSc, Specialist Registrar  R. J. Grimer, FRCS, Consultant Orthopaedic Surgeon  A. Abudu, FRCS, Consultant Orthopaedic Surgeon  S. R. Carter, FRCS, Consultant Orthopaedic Surgeon  R. M. Tillman, FRCS, Consultant Orthopaedic Surgeon  L. Jeys, FRCS, Consultant Orthopaedic Surgeon The Royal Orthopaedic Hospital Oncology Service The Royal Orthopaedic Hospital, Bristol Road South, Birmingham B31 2AP, UK. Correspondence should be sent to Mr R. J. Grimer; e-mail: robert.grimer@nhs.net ©2011 British Editorial Society of Bone and Joint Surgery doi:10.1302/0301-620X.93B7. 25727 $2.00 J Bone Joint Surg [Br] 2011;93-B:980-3. Received 17 August 2010; Accepted after revision 29 March 2011 We investigated the eventual diagnosis in patients referred to a tertiary centre with a possible diagnosis of a primary bone malignancy. We reviewed our database from between 1986 and 2010, during which time 5922 patients referred with a suspicious bone lesion had a confirmed diagnosis. This included bone sarcoma in 2205 patients (37%), benign bone tumour in 1309 (22%), orthopaedic conditions in 992 (17%), metastatic disease in 533 (9%), infection in 289 (5%) and haematological disease in 303 (5%). There was a similar frequency of all diagnoses at different ages except for metastatic disease. Only 0.6% of patients (17 of 2913) under the age of 35 years had metastatic disease compared with 17.1% (516 of 3009) of those over 35 years (p < 0.0001). Of the 17 patients under 35 years with metastatic disease, only four presented with an isolated lesion, had no past history of cancer and were systematically well. Patients under the age of 35 years should have suitable focal imaging (plain radiography, CT or MRI) and simple systemic studies (blood tests and chest radiography). Reduction of the time to biopsy can be achieved by avoiding an unnecessary investigation for a primary tumour to rule out metastatic disease. Bone and soft-tissue sarcomas represent less than 1% of all malignant tumours. 1,2 Delays in diagnosis and a wide range of differential diag- noses are common. 3 These include infections, benign tumours, normal variants, degenerative or metabolic processes and other malignant tumours. Reaching the correct diagnosis in the quickest time possible is essential for the patient and important for resource allocation. 4-7 Current guidelines from the National Insti- tute for Clinical Excellence (NICE) 2 recom- mend that any patient presenting with bone pain or swelling (in particular, pain not aggra- vated by activity) should be referred for radio- graphy. 8 Any suspicious lesion with bone lysis, formation of new bone, periosteal elevation or soft-tissue swelling, should be investigated fur- ther. If a bone sarcoma is considered to be a possibility, the patient should be referred directly to a specialist bone-tumour unit, whereas, if metastasis is suspected, investiga- tion locally is appropriate. 9 For more than 30 years, our centre has accepted referrals from patients who might have a primary malignant bone tumour. Our aim in this study was to analyse the correct diagnosis in all such patients and to define patterns which might lead to clearer guidelines for referral. Patients and Methods We searched our database for all patients referred with a possible primary malignant bone tumour between 1986 and 2010. Patients with a known diagnosis of a bone metastasis or benign tumour were excluded. Most were thought to have had a possible primary malig- nant bone tumour before referral, but in most cases no definite diagnosis had been made and the patients were referred for investigation, diagnosis and advice on treatment. In almost all a ‘suspicious’ plain radiograph had been obtained before referral, but the extent of fur- ther investigations varied widely, with some having had none and others having had CT of the chest and abdomen, bone scanning and MRI of the primary site. All the patients were investigated according to a protocol designed to reach a diagnosis as quickly as possible. In some cases targeted investigations could reach a diagnosis without a biopsy, for example, CT revealing a typical osteoid osteoma. In most, however, the investi- gations included a full history and examination followed by local and systemic imaging. Local imaging was usually CT before 1993 and MRI afterwards. Systemic imaging included bone scanning and chest radiography and in some cases of possible metastatic disease, CT of the