Pathologic fractures can be a source of diagnostic and therapeutic challenge to the practicing orthopaedic surgeon. These often unexpected fractures can be a significant cause of anxiety for patients, who are told they have a tumor, as well as for surgeons. Pathologic fractures, by nature, occur through bone that is biologically abnormal and where the response to and potential for healing can be dramatically different from normal bone for a variety of reasons, including neoplastic and nonneoplastic processes. As a result of this inherent biologic difference, the treatment of pathologic fractures needs to be considered from a different perspective with a different evaluation and treatment algorithm compared to typical fractures.
In these cases, the bone is abnormal for any number of reasons, including metabolic processes that affect the mineralization of bone such as osteoporosis or osteomalacia; medications such as bisphosphonates that suppress bone turnover and affect bone remodeling; treatments such as external beam radiation therapy for treatment of a malignancy; and bone replacement by a neoplasm, which can be either benign or malignant. Benign bone diseases that can predispose a patient to fracture are fibrous dysplasia, unicameral bone cysts, giant cell tumors of bone, and many others. The malignant neoplastic processes encompass diagnoses such as primary bone sarcomas, metastatic bone disease, multiple myeloma, and lymphoma.
Effective and successful treatment of pathologic fractures depends greatly on the reason for which the underlying bone is pathologic, as this has a significant effect on quality of bone stock, potential for healing, and, in the case of malignancy, life expectancy and prognosis.
In the setting of fracture through a benign lesion, an understanding of the natural history of the specific lesion can help to guide treatment. The Enneking staging system for benign tumors can help to provide an understanding for this history as well as guide its treatment. Latent benign lesions (stage I), such as a nonossifying fibroma, can be treated nonsurgically if the fracture pattern allows, as these lesions will heal and regress spontaneously. If the fracture requires surgical stabilization, it can be applied as the fracture pattern dictates, occasionally with concomitant curettage and bone grafting. More biologically active lesions (stage II and III), such as giant cell tumors of bone or aneurysmal bone cysts, require that the surgeon treat both the fracture and the tumor itself. This can be accomplished with immediate treatment of the tumor, often with curettage and stabilization of the fracture; alternatively, the fracture can be allowed to heal nonsurgically and the tumor treated once healing has occurred. Because the bone surrounding these lesions is often normal, healing can occur reliably.
Within the spectrum of fractures through malignant neoplasms, the vast majority are encountered within the setting of metastatic cancers. According to recent data from the Surveillance, Epidemiology, and End Results (SEER) National Cancer Institute database, more than 675,000 cases of lung, breast, and prostate cancer are estimated to be diagnosed in 2013. The rate of metastatic bone disease in these cancers ranges from 30% to 80%. The disease burden, therefore, is significant, and the practicing orthopaedic surgeon will encounter these lesions more commonly than any other.
A fracture through a malignant neoplasm must be treated aggressively. The differentiation between a fracture through a metastatic bone lesion, multiple myeloma, or lymphoma versus one through a primary bone sarcoma must be made before treatment is initiated. In the setting of a primary bone sarcoma, an accurate diagnosis must be established through careful staging and biopsy. These fractures should be stabilized with a cast or minimal internal fixation that can be resected at the time of definitive surgical treatment of the bone sarcoma.
Fracture through bone metastases, multiple myeloma, or a lymphoma of bone are treated with a variety of methods. It is critical to identify the underlying biologic etiology of the fracture because healing rates and treatment options vary widely depending on the type of pathologic bone through which the fracture has occurred. Goals of treatment for impending or actualized pathologic fractures in metastatic disease revolve around palliation and providing the patient sufficient stability of the fracture to allow for immediate, full weight bearing and restoration of function. In these cases, the treating physician cannot rely on the bone to heal, should work under the assumption that local disease progression will occur, and that a construct must be durable enough to last the patient’s lifetime without a need for reoperation. It is important to recognize these fractures and impending fractures and have a treatment strategy for these patients. This chapter will focus on providing a strategy for the evaluation and management of pathologic fractures from metastatic cancers and multiple myeloma.
Metastatic Bone Lesions
Cancer is a major public health problem, causing one in four deaths in the United States today. Breast and prostate cancers continue to be the most commonly diagnosed, although lung cancer continues to be the number one killer in both men and women. Prostate, breast, lung, kidney, and thyroid cancers account for 80% of all skeletal metastasis; and after lungs and liver, the skeleton is the most common site of metastatic disease. The most commonly affected sites are the femur, spine, humerus, pelvis, ribs, and skull, in that order. In regard to breast and prostate cancers, bone is the most common site of metastasis; postmortem examinations show 70% of patients with metastatic bone disease. Carcinomas of the thyroid, kidney, and bronchus have an incidence of 30% to 40% skeletal metastasis at postmortem examination. Furthermore, once tumors metastasize to bone, they usually are incurable. Only 20% of patients with breast cancer are alive at 5 years after the diagnosis of skeletal metastasis.
The exact incidence of bone metastasis is unknown, although it is estimated that 350,000 people with bone metastases die annually in the United States. With the improvement of medical therapies of many cancers, the life expectancy of these patients has increased, which has led to an increasing number of cancer patients surviving with metastatic bone disease. As a result, metastatic bone disease is estimated to cost as much as 17% of the total direct medical costs of cancer treatment in the United States. The consequences of skeletal metastasis are often devastating and can be a major contributor to the deterioration of the quality of life of patients with cancer. Patients can develop severe pain, pathologic fractures, life-threatening hypercalcemia, spinal cord compression, as well as other nerve compression syndromes. Impending and actual pathologic fractures can initiate the period of dependent care for many cancer patients. For all of these reasons, bone metastases are a serious and costly consequence of cancer.
Patient survival after metastasis to bone varies greatly depending on the tumor type and sites of involvement. Mean survival ranges from 6 months, for those with lung carcinoma, to several years, for those with bone metastasis from prostate, thyroid, or breast carcinoma. Also, in breast cancer, prognosis after the development of bone metastasis is considerably better than that after recurrence in visceral sites. Coleman and colleagues found that the median survival of patients with first recurrence of breast cancer in the skeleton to be 24 months, compared to 3 months after relapse in the liver. They also found the probability of survival to be influenced by the development of metastasis at extraosseous sites. Patients with metastatic disease confined to the skeleton had a median survival of 2.1 years, compared to those who later developed extraosseous disease and had a median survival of 1.6 years. When examining lung cancer and skeletal metastasis, Sugiura and colleagues found a mean survival of 9.7 months, with a median survival of 7.2 months. Approximately 70% of patients died within 1 year after skeletal metastasis, and only 6% survived at least 2 years. The mean length of survival was substantially longer in patients with solitary site of metastasis versus patients with multiple sites of disease. The importance of the extent of bone metastasis can also be seen with renal cell metastasis where patients with solitary sites of bone disease treated with wide excision can survive for many years, whereas patients with multiple sites of disease or pathologic fracture have a much shorter survival.
The importance of prognosis for guiding treatment decisions involving these patients cannot be underestimated and many have tried, with little success, to develop models to assist in the decision making for end-of-life orthopaedic care. This information is sought after to help set appropriate expectations for the patient, family, and medical staff. The goal is to maximize function and quality of life for the greatest amount of time. The data about cost, risk, and quality of life are often conflicting, but properly weighed, could help define the most appropriate treatment for an individual with metastatic bone disease.
The decision to pursue surgery, as well as the type of surgery, is strongly influenced by the expected survival of the patient and the need for surgical stabilization. Falsely optimistic survival estimates may influence patients and clinicians to pursue more aggressive therapies, rather than more conservative ones and could result in a higher proportion of both major perioperative complications and death. Conversely, falsely pessimistic survival prognoses could persuade a surgeon to choose a less invasive, less durable fixation technique that lacks sufficient biomechanical durability to outlast the patient.
Nathan and colleagues evaluated patients who had been operated on for pathologic fractures to determine if well-recognized prognostic parameters had any value in determining the survival in this patient population. Median survival in their cohort was 8 months and 60% of the fracture-related consultations to the service underwent operative intervention of both fractures and impending fractures. Independent predictors for survival were diagnosis (or primary site of disease), Eastern Cooperative Oncology Group (ECOG) performance status, number of bone metastases, presence of visceral metastases, and hemoglobin level. Patients with lung cancer fared the worst, and patients with renal cancer fared the best. They concluded, at the end of their study that justification for surgery on the basis of survival prognostication can be dangerously inaccurate and that more accurate prognostic indices are needed for patients undergoing surgery for bone metastases. Forsberg and colleagues attempted to tackle this issue by evaluating three different prognostic models to assist in the decision to offer surgery and also whether a more durable implant was appropriate based on the prediction of 3- and 12-month survival. Ultimately, the treating orthopaedic surgeon makes this difficult decision with input from the rest of the oncology team and careful consideration for the family’s wishes and best interest of the patient.
Biology of Bone Metastases
Metastases can be characterized as either osteolytic or osteoblastic, which represent two extremes within a continuum where the dysregulation of normal bone remodeling occurs. Tumor cells can unbalance coupling in the bone microenvironment leading to bone formation or bone loss. Patients with skeletal metastasis can have either type of lesion or can have mixed blastic/lytic lesions. Specific cancers also have a predilection toward one type or the other. Breast cancer presents with predominately osteolytic lesions, although 15% to 20% can be osteoblastic. In contrast, the lesions in prostate cancer are predominately osteoblastic. Only in multiple myeloma do purely lytic bone lesions develop.
With osteolytic metastasis, the destruction of bone is mediated by osteoclasts rather than tumor cells. Several osteoclastogenic factors have been implicated including interleukin-1, interleukin-6, receptor activator of nuclear factor κB (NF-κB) ligand (RANKL), and macrophage inflammatory protein-1α. Parathyroid hormone-related peptide is also produced by most solid tumors and breast cancer cells and is most likely the factor that stimulates the formation of osteoclasts. The mechanism of formation of osteoblastic metastasis remains unknown, but factors such as endothelin-1, platelet-derived growth factor (PDGF), transforming growth factor-β (TGF-β), urokinase, and prostate-specific antigen (PSA) are thought to be involved.
More than 100 years ago, Paget first noted that the spread of different cancers to distinct organs within the body was not random. He proposed an explanation for this and called it “the seed and soil hypothesis.” The “seed,” the cancer cell, which circulates in the bloodstream, can only “grow,” and metastasize, in particular compatible areas of the body, the “soil.” Not all cancers can grow on all “soil,” which leads to site-selected metastasis.
There are a series of inefficient steps that need to occur for a cancer cell to metastasize. The cell must detach and extravasate from the primary tumor; invade through the extracellular matrix and endothelium to enter the bloodstream; survive within the bloodstream; arrest at a distant site by adhesion to the endothelium; intravasate again through the endothelium and additional extracellular matrix; and finally grow at a distant site. The primary tumor is a heterogeneous population of tumor cells with varying ability to metastasize; some of these cells express prometastatic genes that enable the cells to survive this process and metastasize to bone.
Hematogenous dissemination of cancer to bone is influenced by several factors. Batson described a high-flow, low-pressure, valveless plexus of veins that connects the visceral organs to the spine and pelvis. This vertebral venous system, referred to as “Batson’s plexus,” runs parallel to the vertebral column and forms extensive anastomoses with the venous system of the vertebrae, pelvis, thorax, and brain. However, circulatory anatomy alone does not predict metastasis to bone. Bone receives 5% to 10% of the cardiac output, and as a consequence, most tumor cells that enter the circulation will pass through the bone marrow. Studies comparing perfusion criteria of various organs with metastatic frequency showed no correlation, as there are also other highly vascularized organs to which tumor cells rarely metastasize. Therefore, it is probable that bone provides a particularly fertile microenvironment “soil” for the growth and aggressive behavior of the tumor cells that survive the metastatic process and are able to reach it.
Clinical Features and Presentation
Pain is the most common presenting symptom of metastatic disease to bone. The pathophysiologic mechanism for this pain is poorly understood but includes tumor-induced osteolysis, cytokine and growth factor production, direct infiltration and irritation of endosteal nerve endings, mass effect causing periosteal stretch and elevated intraosseous pressures, stimulation of ion channels, and production of local tissue factors such as endothelin. The pain is usually well localized but can also be a diffuse ache, typically worse at night and not relieved by rest. Eventually the pain worsens with any weight-bearing activity and becomes functional pain. Functional pain is caused by the mechanical weakness of bone that can no longer support the normal stresses of daily activities. Functional pain is typically considered to be an indicator of bone at risk for pathologic fracture.
As a direct result of bone destruction and osteolysis seen with metastatic bone disease, hypercalcemia is a common metabolic complication. One breast cancer study found hypercalcemia in 17% of breast cancer patients with first-time recurrence in bone. Unrecognized, it can be a significant source of morbidity. The signs and symptoms are nonspecific and clinicians should always maintain a high index of suspicion. Mild hypercalcemia may cause unpleasant side effects related to dysfunction of the gastrointestinal tract, kidneys, and central nervous system. As the calcium levels increase, this can lead to renal insufficiency and calcification in the kidneys, skin, blood vessels, lungs, heart, and stomach. Severe hypercalcemia is a medical emergency, and death may ensue as a result of cardiac arrhythmias and renal failure.
Finally, a patient may present with a pathologic fracture; this may be the first sign of metastatic bone disease. Breast, lung, renal, and thyroid cancers are the most common cancers that lead to pathologic fractures. Thirty-five percent of breast cancer patients with metastatic bone disease will sustain a fracture. Patients with bone metastases from prostate cancer usually do not sustain pathologic fractures because of the osteoblastic nature of the disease. However, those with castrate-resistant prostate cancer, where osteoblastic metastasis is typical, annual fracture rates in excess of 20% may be seen.
It is important to understand the differential diagnosis for an adult patient who presents with radiographic findings consistent with an aggressive-appearing skeletal lesion. These include metastatic bone disease, multiple myeloma, lymphoma, primary malignant bone tumor, destructive benign bone lesions, and nonneoplastic conditions (e.g., infection, stress fracture, myositis ossificans, metabolic bone disease, and osteonecrosis). In 2012, there were an estimated 1.64 million new patients with a diagnosis of cancer; of these, it is estimated that greater than 50% are likely to develop bone metastasis. In contrast, only 2890 of these new cases were of patients who presented with primary bone and joint malignancies. Therefore, the chance that a solitary bone lesion is a metastatic carcinoma, especially in an individual older than 40 years of age, is approximately 500 times greater than the chance that it is a primary bone sarcoma. Knowledge of this differential diagnosis helps to guide the diagnostic evaluation of an adult with an aggressive-appearing bone lesion.
Usually, there are three types of patients who are ultimately diagnosed with skeletal metastasis. First is the patient with a remote history of cancer who seeks an opinion regarding an occult or painful osseous lesion. The second has a known cancer history and presents with an asymptomatic skeletal lesion found on routine staging studies. The third type of patient is found to have a skeletal lesion without a prior history of cancer and likely an undiagnosed carcinoma. Regardless of how the patient presents, a thorough history and physical examination is essential to initiate the diagnostic workup. It is important to collect information about current symptoms, cancer history, constitutional symptoms, changes in bowel or bladder function, smoking history, and exposure to chemicals, such as asbestos.
Laboratory studies are part of the diagnostic workup for a patient with a new bone lesion, and although usually not definitively diagnostic, may be helpful and offer clues that will help facilitate staging. Important laboratory values to evaluate include a complete blood count, urinalysis, and chemistry panel. Determination of the erythrocyte sedimentation rate and C-reactive protein are helpful; they are often elevated in individuals with infection, immunologic disorders, or marrow cell neoplasms such as lymphoma or Ewing sarcoma. Metabolic bone diseases such as osteomalacia, hyperparathyroidism, and rickets may be identified with abnormal serum or urine calcium and phosphorus levels. If multiple myeloma is suspected, serum and/or urine protein electrophoresis with immunofixation may confirm this diagnosis. These patients may also have impaired renal function secondary to the presence of Bence Jones proteins. There are also specific blood and tumor markers that can evaluate specific primary sites of disease including thyroid function tests, PSA, carcinoembryonic antigen, α-fetoprotein, β-human chorionic gonadotropin, and cancer antigen–125. These known tumor markers lack specificity and their value usually lies in the assessment of response to therapy more so than in the identification of a primary site of disease.
The clinical imaging evaluation of skeletal metastasis is usually accomplished in one of four ways: plain film radiography, radioisotope scanning, computed tomography, and magnetic resonance imaging. More recently, positron emission tomography (PET) scans have been introduced as another imaging modality to assist in the staging of patients with diagnosed malignant tumors and to evaluate infections and other physiologic processes in the skeleton and soft tissues.
The most important initial imaging modality for evaluation of a bone lesion is a plain radiograph in two planes for any painful lesion. It is important to image the entire bone involved, so as not to miss any discontinuous sites of disease. Plain radiographs can yield more information about a bone tumor than any other diagnostic modality, allowing the clinician to evaluate the anatomic site, the zone of transition between the tumor and the host bone, the internal characteristics of the tumor, and the nature of the matrix that it produces. One can look for aggressive features that include size of the tumor, cortical destruction, periosteal reaction, and pathologic fracture.
Additional three-dimensional imaging of a metastatic bone lesion is typically not needed, unless more precise definition of the soft tissue component is helpful in preparation for surgical or radiation treatment. In these cases, computed tomography (CT) scan or magnetic resonance imaging (MRI) can then be obtained. Once a skeletal metastasis is identified, a bone scan is also typically warranted to evaluate the patient for other bony sites of disease. It should be noted that bone scans identify osteoblastic activity, and disease processes, such as multiple myeloma, with minimal osteoblastic activity, require a skeletal survey for evaluation to prevent falsely negative findings.
PET is an emerging technology that has a high sensitivity for identifying tumors; however, its specificity is quite low. It has been found to be superior to bone scan in detecting bone involvement in various malignancies, and because tracer uptake is not restricted to the skeleton, it has become the mainstay of staging in several malignancies. It can detect lytic, blastic, and mixed lesions because it identifies the presence of tumor directly by measuring its metabolic activity. This method allows for earlier detection of metastatic foci than other studies that indirectly identify tumor by highlighting bone loss as a result of the presence of tumor. It is also much more sensitive than bone scintigraphy, especially in the detection of myeloma or renal cell carcinoma, which are predominately osteolytic. Recent studies have compared PET scans to bone scans and found that PET scans have increased specificity and sensitivity and overall better metastatic lesion detection. However, skeletal scintigraphy, or bone scan, still remains the most commonly used diagnostic imaging modality for the evaluation of the entire skeleton for bony metastases, which is likely due to familiarity with its use compared to the more limited availability and relatively high cost of PET scans. However, with the recent surge in interest, gradually increasing availability, and increasing spectrum of applications for PET scans, this is rapidly changing and newer imaging and treatment modalities are constantly evolving.
Understanding that many patients with skeletal metastasis are older than 40 years of age, present with a destructive painful bone lesion, and have an unknown primary, Rougraff and colleagues developed a diagnostic protocol to assist the orthopaedic surgeon who will have the task of determining the primary site of malignancy. By obtaining an adequate history and physical, routine laboratory analysis, plain radiographs of the involved bone and chest, whole body bone scan, and CT scan of the chest, abdomen, and pelvis with oral and intravenous (IV) contrast, they were able to identify the primary tumor site in 85% of patients.
A biopsy is performed only once all of the workup has been done and the data necessary to assist in the diagnosis has been collected. There are several reasons why it is critical to conduct a staging workup prior to the biopsy. First, the tumor may be a primary sarcoma of bone and an ill-planned biopsy could compromise the ability to perform a limb-sparing procedure and to obtain high-quality imaging studies. Second, there may be another site of disease that is easier to biopsy and associated with less morbidity. Third, preoperative embolization may be helpful to prevent bleeding during biopsy or treatment, such as in the case of presumed renal cell metastases. Fourth, an unnecessary biopsy can be avoided if the diagnosis can be made based on laboratory analysis alone, such as with multiple myeloma. Fifth, histologic analysis alone identified the primary site of disease in only 3% of patients. Sixth, the increased information, as a result of combining laboratory studies with imaging results and histopathology, make it more likely that an accurate diagnosis is obtained.
There are multiple ways to perform a biopsy and specific guidelines that must be adhered to. Biopsy techniques include fine-needle aspiration, image-guided core-needle biopsy, or open incisional biopsy. Proper oncologic principles should always be followed. The advantage of fine-needle aspirations or core-needle biopsies is that general anesthesia is not required, they are less morbid procedures, and they reduce the potential for contamination of the tumor site. The main disadvantage is a potential for sampling error due to the smaller tissue sample resulting in lower rates of accurate diagnosis compared with open biopsy. When performing an open incisional biopsy, the incision should be as small as possible, should be oriented in a longitudinal fashion with minimal disruption of the surrounding tissues, and should avoid major neurovascular structures. The location of the biopsy must be chosen cautiously, so it can be excised en bloc with the tumor, if necessary. It is important to maintain hemostasis throughout the procedure using electrocautery and bone wax to minimize contamination and tumor cell spillage.
The majority of metastatic bone lesions are treated effectively with nonsurgical modalities such as radiation therapy, chemotherapy, immunotherapy, hormonal therapy, and bisphosphonates. Operative treatments are palliative procedures with the goals of achieving local tumor control and structural stability, allowing the patient immediate function and weight bearing with the least possible morbidity and need for rehabilitation. The question is always, when to operate? To help with this decision, many physicians rely on Harrington’s classic definitions of impending pathologic fractures of the long bones or the Mirels classification system. Harrington’s criteria for impending pathologic fracture includes a bone lesion that measure 2.5 cm or greater in the proximal femur, occupies 50% or more of the bone diameter, is accompanied by an adjacent lesser trochanter fracture, and has not responded to treatment with radiotherapy. However, subsequent studies have failed to prove a strong relationship between these factors and risk of fracture.
The Mirels scoring system is based on four parameters: site, radiographic appearance, size, and related pain. Mirels reviewed 78 patients with radiated lesions of the long bone and devised a scoring system to predict the risk of pathologic fracture. This numerical scoring system has a maximum of 12 points. For scores greater than 8, prophylactic stabilization of impending pathologic fracture was recommended, while for scores of 7 or less, nonsurgical treatment was recommended due to the low fracture risk. For a score of 8, the decision must be individualized. Mirels’ scoring system has the advantage of being simple, reproducible, and valid across experience levels. The disadvantages are that it has never been evaluated prospectively and, although highly sensitive as a screening tool, it has relatively poor specificity in predicting actual fracture.
Unfortunately, impending and actual pathologic fractures often initiate the period of dependent care for many cancer patients, which is why this decision is of the utmost importance. Identifying an impending fracture and recommending prophylactic fixation is an important issue. Elective fixation prevents the pain and suffering associated with a pathologic fracture, and prophylactic fixation is often easier to perform. Ward and colleagues noted in a retrospective, nonrandomized study, that treatment of impending pathologic fractures yielded better results than treatment of complete fractures. There was less average blood loss, shorter postoperative hospital stay, greater likelihood of discharge to home as opposed to an extended care facility, and a greater likelihood of resuming support-free ambulation. Patients with impending fractures also fared better in terms of survival at 1 and 2 years.
The treatment of bone lesions before or after fracture depends on the underlying lesion and its location. If the suspicion is high for a primary bone tumor, whether benign or malignant, it is best to refer to an orthopaedic oncologist. Common benign lesions include nonossifying fibromas, bone infarcts, unicameral bone cysts, fibrous dysplasia, aneurysmal bone cysts, and giant cell tumors of bone. The decision to observe, curettage and graft, and/or internally fix should be determined by someone familiar with these uncommon entities. Malignant primary bone tumors, such as osteosarcomas or chondrosarcomas, benefit from multidiscipline care, which may include chemotherapy, radiation, surgical resection, and soft tissue coverage procedures. The modern orthopaedic oncologist is able to pursue curative intent with limb salvage in 95% of cases. Pathologic fractures sustained in the setting of a primary bone sarcoma are no longer an immediate indication for amputation because of the effectiveness of modern-day systemic therapies. Consequently, many orthopaedic oncologists will splint or perform limited internal fixation of these pathologic fractures and wait to see how patients respond to neoadjuvant treatment before deciding on definitive surgical management. Secondary bone lesions from metastatic carcinoma and the primary bone marrow malignancies are the most common cause of pathologic fractures. While specialized care for these lesions is also appropriate, they can also be appropriately treated at nonspecialty centers, if basic principles are acknowledged.
The most important first step is establishing the underlying cause of a pathologic fracture. There can be many reasons for bone to be weak such as osteoporosis, prior radiation, medication use, metabolic derangements, infection, and malignancy from, or traveling to, bone. If there is any uncertainty, an open biopsy must be performed first. Principles of biopsy were discussed earlier. It is not acceptable to send intramedullary reamings to establish diagnosis because the whole bone has already been violated if the lesion turns out to be a primary sarcoma ( Fig. 20-1 ). In the setting of a known metastatic tumor, it may be appropriate to send reamings in order to assess tumor response to treatment. Treatment varies depending on the diagnosis of the primary tumor. Primary bone lesions and solitary metastatic tumors from thyroid and renal cell carcinoma metastatic lesions to bone are sometimes treated with local surgical resection and curative intent. Other secondary bone lesions from carcinoma and widespread bone marrow malignancies are treated with palliative intent. The surgical goals are unique for these patients and the principles of fracture management differ from conventional management of long bone fractures.
The vast majority of patients with metastatic disease to bone, with the uncommon exception of isolated lesions from thyroid or renal cell primaries, are not going to be cured through surgery. The orthopaedic surgeon needs to view these patients and goals of surgery differently. First, quality of life and the maintenance of function are the primary goals, not the prolongation of life. Surgical strategies to allow immediate, full weight bearing, minimize pain, facilitate mobility, and minimize time in the hospital are the priority. Second, the benefit of surgery and time to recover should be weighed against the expected patient survival. If the latter is less than the recovery time, a less invasive strategy to address pain should be planned. Third, pathologic bone is weak, of poor quality, and cannot be relied upon to heal. Further, additional measures in the form of chemoradiation can worsen the healing potential. At best, healing is significantly delayed, often taking longer than 6 months. Fixation needs to be durable, tolerate a greater load for a longer period of time, allow immediate full weight bearing, outlive the patient, and avoid reoperation. Fourth, endoprosthetic reconstruction is a consideration for initial management ( Fig. 20-2 ). Extensive pathologic bone, likely disease progression in treatment-resistant tumors, and limited alternatives that could allow immediate weight bearing are all instances in which to consider an endoprosthesis as first-line treatment. Fifth, there must always be an attempt at local control of the tumor. This can be from complete wide resection, curettage, systemic therapies, or with external beam radiation therapy (EBRT). Radiation is most commonly used after surgery and the field frequently encompasses the entire length of the surgical implant. Sixth, these patients need careful soft tissue management. Providing a setting for uncomplicated healing after surgery facilitates early adjuvant treatment, prevents wound breakdown from radiation, and protects against infection when patients are immunosuppressed. Consideration of these factors goes a long way in keeping patients out of the hospital and improving quality of life.
Determining the best treatment for metastatic disease requires knowledge of both the tumor biology and the surgical technique. Life expectancy is more favorable in those with solitary bone lesions, without visceral metastases, and in those with treatment-responsive subtypes, such as prostate, breast, and multiple myeloma. The capacity to heal should be considered in these patients. Gainor and colleagues showed that 60% to 70% of pathologic fractures from these histologic subtypes will heal in longer than 6 months. Internal fixation enhanced healing, whereas adjuvant radiation did not appear to have an influence. On the contrary, lung cancers and melanoma have poor prognoses and are notoriously resistant to systemic treatments and radiation. These patients have shorter life expectancies and disease that is likely to progress. Early aggressive surgery is favored in those medically fit for such intervention, as healing cannot be relied upon. Renal cell carcinoma has traditionally been associated with a poor prognosis. However, new targeted therapies that inhibit angiogenesis have improved disease control and patient longevity. Local management of renal cancer metastatic to bone remains a challenge and will be discussed later in additional detail.
In all these patients, a few simple surgical techniques should be considered. As noted earlier, implants should be expected to bear a greater load for a longer period of time and should provide immediate axial and rotational stability. Therefore, long, load-sharing implants are preferred when possible. They are durable and can prophylactically protect long segments of bone to avoid future fractures in the event of disease progression ( Fig. 20-3 ). These devices are strongest when made from stainless steel but this strength must be weighed against the need for future imaging with MRI, where a titanium implant may be more appropriate. Significant tumor load may benefit from open debulking to decrease the local disease burden and enhance the effectiveness of adjuvant radiation for local tumor control. The remaining defect(s) can be supplemented with polymethylmethacrylate (PMMA) to improve the rigidity of the construct and facilitate early weight bearing. PMMA is not weakened by radiation and, when used to fill cavities, does not prevent bone healing.
Amputation is rarely done for pathologic fractures from metastatic disease. It is primarily considered in tumors that progress and ulcerate through the skin despite treatment, painful nonunions after failed fixation without other reconstruction options, extensive bone loss or neurovascular involvement, and for postradiation changes that result in a painful, nonfunctioning extremity.
Although pathologic fractures are nearly all indicated for operative intervention, certain fractures in treatment-sensitive malignancies can be treated without surgery. Fractures of the upper extremity, ribs, and clavicle are nonweight bearing areas that can be well managed with casting, bracing, or pain-control measures. Patients can remain quite functional and have a favorable quality of life during treatment. Humeral shaft fractures are particularly well tolerated despite the long course of treatment with functional bracing ( Fig. 20-4 ), and many pathologic vertebral compression fractures can be treated nonsurgically with good results. Small, nondisplaced fractures in low stress areas of the lower extremities or hard-to-reach areas of the pelvis, such as the ischium, pubis, and sacroiliac joint, can be considered in patients with favorable life expectancies as long as their pain is mild, cortical destruction is minimal, and they are willing to adhere to weight-bearing restrictions while receiving radiation or systemic therapies ( Fig. 20-5 ). Close follow-up is necessary to monitor the response to treatment and assure patient function and quality of life is not being compromised.
There are times where surgery is indicated but the poor health status of patients renders them medically unfit for an operative procedure, which is usually because of extensive comorbidities and/or advanced state of the malignancy leading to a poor life expectancy. Comfort measures become the priority with a focus on local disease control and pain management.
Local tumor control without surgery puts the focus on adjuvant treatments. Chemotherapy is an option for all metastatic cancers and is specific to each tumor type. The systemic action leads to diminished tumor load and a decrease in musculoskeletal pain. Lymphoma is particularly sensitive to chemotherapy and often does not require any additional treatments even when bone is involved. Use of chemotherapy is limited by the severity of toxicity. As a result, elderly and medically complicated patients are often not candidates. Hormone therapy is an option in breast and prostate cancers. Agents for breast cancer include estrogen-receptor antagonists and aromatase inhibitors, which are effective in around 50% of patients. Prostate cancer treatments include gonadotropin-releasing hormone (GnRH) analogs, cytochrome P450 inhibitors, androgen antagonists, and 5α-reductase inhibitors. They decrease bone pain from prostate cancer metastases in 70% to 80% of patients.
Another strategy is to target the tumor activation of host osteoclasts that leads to bony destruction. Bisphosphonates and RANKL inhibitors are used to combat this process and have been shown to decrease bone destruction and the incidence of pathologic fracture. Bisphosphonates work by binding bone mineral and being internalized by osteoclasts and ultimately induce apoptosis. They are effective in preventing progression in both osteolytic and osteoblastic bone lesions. Although best studied in breast cancer and multiple myeloma, they have proven effective across a wide range of cancers. Zoledronic acid (Zometa, Novartis, Basel, Switzerland) has proven to be effective in achieving pain relief, decreasing the need for radiation, preventing hypercalcemia, and preserving function. Administered as a 4-mg IV infusion every 3 weeks, skeletal-related events (SREs), defined as pathologic fracture, surgery for bone complication, use of radiation, or hypercalcemia, can be diminished by 35% and the time to the first SRE is delayed by 70 days. There may be some direct antitumor effect from bisphosphonates as well, but this has been debated. Denosumab (Xgeva, Amgen, Thousand Oaks, CA) is another medication able to interfere with tumor activation of host osteoclasts. It is a human immunoglobulin G (IgG) monoclonal antibody to RANKL and prevents the activation of RANK receptors on osteoclasts. Given as a 120-mg subcutaneous injection once a month, it has proven more effective than bisphosphonates in the reduction of SREs and time to first SRE. There is no difference between the two medications in regard to disease progression or overall patient survival.
Direct local control in impending and pathologic fractures not treated with surgery either due to location, minimal symptoms, or patient comorbidities is most commonly treated with EBRT. Radiation, administered in single or multiple fractionated doses, achieves complete pain relief in around 30% and partial relief in another 30% to 40%. The onset of relief takes 2 to 4 weeks and the average duration is 6 months. A single fraction of 8 Gy is as effective as multifraction radiation for pain relief but has a higher incidence of retreatment. Multifractional treatment is generally preferred for neuropathic pain and in patients with longer life expectancies. EBRT for palliation has the additional benefit of assisting bone repair by interfering with osteoclasts and ossifying collagen strands from the proliferative fibrous tissue that replaces metastatic cells. Use can decrease the risk of reoperation from 15% to 20%. Stereotactic radiation utilizing steep dose gradients to protect vital structures has been adopted from the treatment of brain malignancies to palliate disease to the spine ( Fig. 20-6 ).
Various techniques exist to address metastatic lesions through percutaneous or endovascular techniques. Each has specific indications and is useful in isolation or as an adjunct to surgery. Radiofrequency ablation (RFA) is considered in small (≤5 cm), contained lesions that are painful. In management of pathologic fractures, this technique is primarily used in pelvic lesions with associated insufficiency fractures ( Fig. 20-7 ). Performed percutaneously with CT, ultrasound, or fluoroscopic image-guidance, heat is generated within the lesion to directly ablate tumor as well destroy local sensory nerves. This ablation is often followed by injection of PMMA to provide increased structural stability at the tumor site. RFA can spare exposure to EBRT and is effective, with 95% achieving pain relief lasting for 6 months.
Administration of radioisotopes is another technique used to address symptoms from multifocal metastatic lesions too numerous for EBRT and resistant to other systemic medical therapies. The radioisotopes, given as a single outpatient injection, include strontium-89, phosphorus-32, samarium-153, and radium-223, and are most effective in prostate and breast carcinomas. The isotopes have an affinity for bone and emit radioactive particles in a range of 0.2 to 3 mm, which helps localize the radiation and minimize side effects, which are primarily myelosuppression and thrombocytopenia. Treatment is contraindicated in those with a poor performance status, less than 2-month survival, extensive soft tissue metastases, a platelet count below 60 × 10 3 /µL, recent disseminated intravascular coagulation, impending fractures, and cord compression. Most patients respond in 2 to 4 weeks and 60% to 80% achieve modest pain relief that lasts longer than 6 months. Radioisotopes have been associated with decreased use of pain medications, less need for EBRT, and the development of fewer new bone metastases. Retreatment can occur every 10 weeks. Bone marrow recovery must occur after administration, which limits use of this modality in conjunction with EBRT and chemotherapy.
Embolization is another modality that can be utilized on its own to address difficult-to-reach anatomic areas or in preparation for surgery to decrease blood loss. Using percutaneous endovascular techniques, interventional radiologists use polyvinyl alcohol, coils, or gel foam to occlude large feeding vessels to the tumor. Pathologic fractures secondary to thyroid, renal, lung, and myeloma malignancies should receive preoperative embolization to reduce blood loss ( Fig. 20-8 ). Surgery should take place within 48 to 96 hours, before revascularization occurs.
Orthopaedic intervention for impending and pathologic fractures decreases pain and improves mobility in 90% of patients. As a result, surgical stabilization is recommended unless patients are too sick or disease is too progressive with a poor life expectancy. Surgical approach and recommended fixation varies by location and therefore, our discussion will consider specific anatomic areas. It is important to note that most surgically treated impending and pathologic fractures are treated postoperatively with EBRT within 2 to 3 weeks of surgery, once wounds have healed. Further, almost all patients are made weight bearing as tolerated immediately after surgery.
Twenty percent of metastatic lesions to bone occur in the upper extremities. Impending and pathologic fractures occur most commonly in the proximal humerus and humeral shaft. Distal lesions can occur and often result from lung and renal primaries. While less devastating than fractures in the lower extremity due to lower stress and limited need for weight bearing, patients have improved pain control and maintained function after operative stabilization of pathologic fractures. The mode of fixation is dependent on the location of fracture and extent of destruction.
Fractures in the humeral head and anatomic neck are indicated for a cemented hemiarthroplasty ( Fig. 20-9 ). The length of the stem should bypass all lesions in the humerus, ideally by two cortical diameters. It is imperative to obtain radiographs of the entire long bone before determining the length of the stem. Lesions on the glenoid side can be treated with curettage and cementation with or without resurfacing. An active patient with a rotator cuff deficient shoulder with an intact deltoid warrants consideration for a cemented reverse total shoulder to maximize abduction and forward flexion. Historically, hemiarthroplasty has been favored over total and reverse shoulder arthroplasty due to reliable pain relief and stability. Function, however, was rather poor. A growing focus on patient-derived outcomes has pushed thinking toward techniques with better functional outcomes. Because resections for metastatic lesions are intralesional, surgeons are able to maintain more structures to aid stability, such as the joint capsule and rotator cuff tendons, improving stability in total and reverse shoulder arthroplasties. When used for proximal humerus malignancies and at a mean follow-up of 7.7 years, the reverse total shoulder provided a mean abduction of 157 degrees and significantly improved activities of daily living. Advanced disease in the proximal humerus with a large soft tissue mass and extensive bone loss is a reason to consider an endoprosthesis ( Fig. 20-10 ). If remaining diaphyseal segment is very short, options include a compliant prestress fixation-type endoprosthesis, shortening the stem with a high-speed cutting burr on a conventional endoprosthesis, or a total humeral replacement.