In 2005, cancer surpassed cardiovascular disease as the leading cause of death in the United States. Metastatic bone disease represents an increasing burden on the healthcare system and requires special consideration by orthopaedic surgeons. Advances in systemic therapy are prolonging survival, which could lead to an increased prevalence of metastatic disease. Furthermore, an aging population and stable disease incidence are factors that suggest that the number of patients with advanced disease will increase. The management of skeletal metastasis requires consideration of patient-specific and disease-specific factors combined with a systematic orthopaedic approach.

Ultimately, the decision for surgical intervention in the patient with metastatic bone disease is a collaborative one. The individual patient, their overall health, and the severity of disease must be taken into account. In the management of metastatic cancer, the orthopaedic surgeon is just one member of a comprehensive treatment team, and optimal care and outcomes depend on the multidisciplinary efforts of that team.

Traditionally, patients scheduled for orthopaedic procedures undergo medical optimization, or medical clearance in the case of more urgent procedures. One of the primary tools used by internists for standard risk stratification is the American College of Cardiology/
American Heart Association Guidelines on Perioperative Cardiovascular Evaluation;¹ however, these guidelines have significant limitations regarding the management of impending or actual pathologic fractures. Specifically, they only address perioperative cardiovascular risk and may not accurately portray the scope of the planned intervention, as all orthopaedic procedures are considered intermediate risk.

The American College of Surgeons has developed a more comprehensive online tool using the National Surgical Quality Improvement Program database for a variety of postoperative complications based on the type of surgery, histology, and patient factors.^2,3 Figure 1 shows a segment of the results generated by this calculator, which is available online. In addition, the Timed Up and Go test initially developed for evaluating patients with hip arthroplasty has shown utility in predicting outcomes in oncologic surgery. A Timed Up and Go test score higher than 20 seconds is an independent risk factor for postoperative complications and is twice as sensitive compared with the American Society of Anesthesiologists score.⁴

Medical optimization should include the correction of modifiable factors commonly seen in patients with metastatic bone disease. Laboratory aberrations may be the result of the nature of the disease (ie, renal failure in myeloma) or secondary to systemic treatment such as chemotherapy. Two important considerations are hypercalcemia and bone marrow suppression resulting in anemia, thrombocytopenia, and immunosuppression. Hypercalcemia is seen in up to 20% of patients with cancer and can cause significant morbidity and death. Patients with a serum calcium level higher than 14 mg/dL should be treated expeditiously with hydration and diuretics.⁵ Preoperative anemia (hemoglobin level <7 g/dL) is associated with a poorer prognosis⁶ and transfusion should be considered before any procedure. Neutropenia is defined as an absolute neutrophil count less than 1,500 cells/µL, and less than 500 cells/µL is considered severe neutropenia.⁷ Some authors suggest an absolute neutrophil count of at least 1,000 cells/µL before surgical intervention is considered.⁸

With regard to surgical site infection, independent risk factors include neutropenia, preoperative radiation, and metastatic cancer.⁹

In the case of a pathologic fracture, the urgency and necessity of surgical treatment are amplified. Although
there is no definitive cutoff for intervention, the quality of life and palliative effect of surgery must be weighed along with patient and family wishes and estimated survival. Surgery for metastatic bone disease is usually palliative, and one of the main tenets is maintaining or improving quality of life.¹⁰ Fractures of the lower extremity cause significant morbidity in addition to pain and loss of independence. It was shown that most patients in hospice care (83.4%) who elected to undergo surgical treatment of hip fractures had improved survival.¹¹ Because upper extremity fractures do not require the same urgency as fractures of the lower extremity, many of these patients can be treated with immobilization or treatment deferred because they do not carry the same risk of becoming bedbound or being hospitalized.¹²

Although no definitive or perfect method for the determination of survival in any particular patient exists, some useful tools are available that may be used. The Palliative Prognostic Score is a validated tool for predicting 30-day mortality.¹³ PATHFx 3.0 is an online tool (www.pathfx.org) that can be quickly used to determine prognosis in the setting of impending or completed pathologic fracture. It takes into account histology, patient factors, completed or impending fracture, and laboratory values. The scoring system proposed in 2014 accounts for some of these same variables and stratifies patients based on a prognostic score.¹⁴ Although overall survival is generally poor (Figure 2, A), it can vary substantially (Figure 2, B) and thus may affect decisions regarding treatment.

One of the most important disease-specific factors to consider is histology. Histologic diagnosis not only affects outcome (ie, survival) but also influences management choices (Figure 2, B). The histologic diagnosis goes well beyond predicting survival and influences a range of findings in metastatic lesions, such as vascularity (ie, need for preoperative embolization), risk of local progression, response to systemic therapy or radiation, and overall prognosis. An identical femoral lesion from renal cell cancer may require treatment different from one that causes hormone receptor-positive breast cancer.

Despite the best clinical judgment and all of the available tools for predicting survival, certain patients have shown unpredictable outcomes. The radiographs in Figure 3 are from a patient with lung cancer metastatic to bone. Usually, this diagnosis is associated with very poor survival; however, after initial treatment with femoral nailing, the patient lived for more than 1 decade and underwent numerous procedures to treat nonunion, hardware failure, recurrent pathologic fracture, and infection. Because patients may survive longer than predicted, it is important to make sure that the construct is as durable as possible. As the efficacy of medical therapies improves, it will become more important to provide long-lasting fixation.

Solitary bone metastasis and oligometastatic disease present a particular challenge, but also potentially represent an opportunity for improved outcomes. There is no standard definition of oligometastasis in terms of either number or location; for bony disease, this is generally considered to be up to four sites. Similarly, a single bone lesion and a single hepatic lesion would also be considered an oligometastatic state. When considering treatment of patients with this condition, it may be more helpful to consider oligometastatic disease as one in which all sites of macroscopic disease could potentially be surgically resected.

In this context, it may be preferable to consider a resection rather than an intralesional procedure. Resection may be particularly attractive in certain situations: isolated bone metastasis in a patient with a relatively good prognosis; a tumor with high likelihood of local progression and poor response to systemic therapy; metastasis in patients with a long disease-free interval; or metastasis to expendable bones, such as the fibula, ribs, clavicle, and certain spinal elements.

Although cure of metastatic disease is the exception rather than the rule, there seems to be a survival advantage with en bloc resection compared with intralesional procedures^15,16 (Figure 4). In the proximal femur, for example, endoprosthetic replacement is associated with a survival advantage and improved functional scores compared with internal fixation.¹⁶ Although there are scoring systems that predict outcomes of resection to manage metastatic disease in the liver,¹⁷ no such system exists for bone.

Assessing the risk of fracture has been a long-standing challenge for metastatic lesions and the subject of considerable research. Pathologic fracture is associated with pain, expense, loss of independence, and decreased survival rate for these patients⁶ (Figure 5).

In past decades, the Mirels criteria have been used to predict risk of pathologic fracture based on location, pain, and attributes of the lesion on plain radiographs.¹⁸ However, these criteria are subjective and lack reproducibility in predicting fracture risk.¹⁹ Furthermore, the Mirels system was developed before the advent of bisphosphonate (also known as diphosphonate) therapy and contemporary chemotherapy and does not take into account the rapid tumor response, which may be observed in some patients with targeted therapy. This recognition shifts the decision making in surgical intervention for patients, because surgical complications that delay commencement of effective chemotherapy may actually negatively affect survival. Furthermore, rapid response may obviate the need for fixation in selected patients. As such, the orthopaedic surgeon should also take into account the patient and tumor characteristics in combination with symptoms and imaging. The decision to recommend surgical intervention is based not only on the risk of fracture but also on a careful discussion with the patient and the care team regarding the goals of care and risk-benefit ratio. In the contemporary era, the decision for prophylactic fixation should also be made in concert with the medical oncology team and with recognition of the role that surgery can play as evaluated in the entire context of the patient’s care.

As an alternative to the Mirels criteria, which may overestimate fracture risk, early results of patient-specific risk of fracture based on CT are promising.²⁰ A finite analysis model was shown to be superior to experienced clinicians at identifying patients at a high risk of fracture.²¹ This technique has many limitations and has not been proven on a large scale, but as computer modeling and imaging technology improve, this may be a valuable addition to help in the prediction of fracture risk.

The Tokuhashi Scoring System was developed in 1990 to predict survival in patients with metastatic spine disease.²² This scoring system was later revised to rely more heavily on the histology of the primary tumor in predicting outcome.

The Revised Tokuhashi Scoring System has been shown to predict prognosis in more than 80% of patients with metastatic spine disease and has been validated as a useful tool in selecting an appropriate treatment modality.²³ For patients with a prognosis of less than 6 months or multiple involved vertebrae, nonsurgical treatment or palliative surgery was elected, whereas excisional surgery was performed in patients with a prognosis longer than 1 year, or longer than 6 months with metastasis in a single vertebra.