C-reactive protein and tumour diagnosis predict survival in patients treated surgically for long bone metastases

Purpose: Surgical options for long bone metastases include intramedullary nail fixation or prosthetic reconstruction. Patients with a short life expectancy may benefit from less invasive surgery such as intramedullary nail fixation, while patients with a long life expectancy could be treated with more invasive surgery such as prosthetic reconstruction. The purpose of our study was to analyze the survival of patients treated surgically for long bone metastases, determining the prognostic factors affecting survival and analyzing the surgical complications and reoperation rates. Based on our results, we developed a prognostic score that helps to choose the best treatment for these patients. In addition, we compared the performance of our prognostic score with other previous prognostic models. Method: We investigated prospectively potential clinical and laboratory prognostic factors in 159 patients with metastatic bone disease who underwent surgery with intramedullary nail fixation or prosthetic reconstruction. Clinical data were collected, recording the following data: age and sex of patients, primary tumour and time of diagnosis, number (single or multiple) and presentation (synchronous or metachronous) of bone metastases, presence of visceral metastases. The following laboratory data were analyzed: hemoglobin, leukocyte counts, lymphocyte counts, platelets count, alkaline phosphatase, and C-reactive protein. Results: Our study showed that pathological C-reactive protein and primary tumour diagnosis were significant negative independent prognostic factors at 12-month survival. Based on our results, we created a score using C-reactive protein and primary tumour diagnosis, creating three different prognostic groups: (A) good prognosis primary tumour and physiological CRP with probability of survival at 12 months of 88.9 [80.1–98.5]; (B) bad prognosis primary tumour and physiological CRP or good prognosis primary tumour and pathological CRP with a probability of survival at 12 months of 56.7 [45.4–70.7]; (C) bad prognosis primary tumour and pathological CRP with a probability of survival at 12 months of 12.5 [5.0–28.3]. Using ROC multiple analysis, our score (AUC = 0.816) was the most accurate in predicting a 12-month survival compared to previous prognostic models. Discussion: Patients treated surgically for long bone metastases with a life expectancy over 12 months should be treated with more durable reconstruction, while patients with a life expectancy less than 12 months should be treated with less invasive surgery. The diagnosis of primary cancer and C-reactive protein are two very simple data which every orthopaedic surgeon in any community hospital can easily rely on for any decision-making in the surgical treatment of a complex patient as with a patient with skeletal metastases. Conclusion: Our prognostic score based on only two simple variables (C-reactive protein and primary tumour diagnosis) was able to predict the 12-month survival of patients treated surgically for long bone metastases and could be helpful in choosing the best treatment for these patients.