Bone scintigraphy in lung cancer: a reappraisal

The prognostic significance of bone scintigraphy was investigated by following 587 consecutive patients with lung cancer, in whom this investigation had been performed, for up to 9 years, or until death. Survival was unrelated to age, sex or cell type. However, pain and abnormal bone scintigraphy were both independently associated with a significantly reduced survival compared with those who were free of pain or who had normal bone scintigraphy. These factors were cumulative. The association remained equally valid for all cell types. Claims that a single metastasis is not prognostically significant are unfounded. It is suggested that the results of some chemotherapy trials must be reconsidered in the light of present findings, because of the lack of adequate control groups; the results could be construed to show a beneficial effect only in patients with bone metastases and a poor prognosis, but little or no effect in patients with normal bone scintigraphy. As judged by clinical and radiological follow-up and post-mortem examination, skeletal scintigraphy in patients with lung cancer had a sensitivity of 0.89, a non-specificity (false positives/true negatives) of 0.00 and an accuracy of 0.78. With existing radiopharmaceuticals there is an irreducible residue of false negatives due to deposits which provoke little or no osteoblastic response. Bone scintigraphy is, thus, indicated in any patient with lung cancer with unexplained symptoms and whenever staging is required, because of the prognostic implications. It should precede other staging investigations because the high detection rate may render other tests unnecessary.