Surgical Approaches to Thoracic Primary and Secondary Tumors




Overview


The past four decades have witnessed major advances in spine surgery, and these pioneering developments have dramatically influenced the culture and attitudes surrounding surgical management of spine tumors. At one point considered controversial, surgical intervention for spine tumors is now considered the primary management modality for many tumors, and such intervention has significant advantages over medical management. The safety and efficacy of such advancements has recently been underscored in several key publications with respect to three major outcomes: 1) spinal cord compression, 2) pain relief, and 3) improvement in quality of life. In most cases, the goals of surgery are to offer decompression and stabilization and, as a consequence, to provide tremendous palliative support to the patient. In fact, in more unique circumstances, such as that of rare primary tumors of the spine or solitary metastases, surgery may actually be curative. There is no doubt these breakthroughs have created an exciting landscape in spine surgery, and particularly in the thoracic spine, more so than any other area of the spine, these advances have converged to make the complex management of tumors truly a multidisciplinary enterprise.




Incidence of Primary and Secondary Tumors of the Thoracic Spine


Current studies estimate that about 50% to 70% of all cancer patients have metastases at the time of their death. Osseous lesions represent a significant disease burden in this population, and the axial spine has the highest predilection to such lesions. The primary site of malignancy for the majority of these spine lesions most often includes the prostate, breast, skin, and lung. Remarkably, the aggressive biology of these neoplastic lesions leads to pathologic fractures that often result in epidural spinal cord compression; the majority of these events arise from direct metastases from breast (22%), lung (15%), and prostate (10%) cancers. Of all the patients who develop metastatic spinal disease, approximately 20% will develop pathologic fractures that cause thecal sac compression, which accounts for the symptomatic presentation of nearly 20,000 to 30,000 patients annually for such cases. The majority of secondary tumors of the spine will often present in the thoracic spine, (70%) with the remainder presenting in either the lumbar (20%) or cervical spine (10%). A recent study has delineated the precise involvement of the vertebral elements and surmises that the majority of these lesions often present with involvement initially of the posterior half of the vertebral body, but with time, the lesion invades into the anterior body, lamina, and pedicles.


Primary tumors of the vertebral column are rare neoplastic lesions, but they account for approximately 10% of all primary osseous tumors, which culminate in 7500 new spinal column tumor diagnoses each year. This number is dwarfed by the nearly 90,000 new cases of metastatic axial spine tumors diagnosed annually, and it quickly puts into perspective the incidence of primary versus secondary tumors of the spine. Despite this, primary tumors of the axial spine should be strongly considered in the differential diagnosis of spinal lesions, because major advances in the excision of these tumors may undoubtedly influence the prognosis and surgical approach.


The majority of primary tumors of the spine can be classified as either benign or malignant. Benign primary spinal tumors include aneurysmal bone cysts, heman­giomas, osteoid osteomas, osteoblastomas, osteochondromas, enchochondromas, chondroblastomas, and giant cell tumors. Malignant spinal tumors include chordomas, chondrosarcomas, Ewing sarcomas, osteosarcomas, plasmacytomas, and multiple myelomas. In contrast to the management of secondary tumors of the spine, the major goal of surgical intervention for primary tumors is to achieve local tumor control to influence survival and possibly relieve the patient of the total tumor burden. Currently the most accepted and advanced surgical intervention for these tumors includes en bloc resection with wide excision margins for invasive, benign primary spinal tumors and nearly all malignant primary spinal tumors not responsive to chemotherapy or radiation.




Clinical Presentation and Evaluation


Despite the broad spectrum of pathology that can exist among various primary and secondary tumors of the spine, a single unifying feature between them is that pain is central to their initial presentation. Back pain is often exhibited by patients well before neurologic deficits manifest. This pain is often characterized as either mechanical or biologic (tumor related). Biologic pain is associated with advanced tumor burden and is often insidious and dull in character, persistent in nature, and most often reported to be worse at night (nocturnal pain) or early in the morning. For clinicians, this pain can be distinguished from muscular or inflammatory pain because this pain is often described as unremitting, exacerbated in the supine position, and often less relieved by positional changes; it is most notably not responsive to analgesics. The pathophysiology of biologic pain is most often mechanistically associated with invasive margins destroying one or more of the facet joints. Several other mechanisms have been proposed that include local release of cytokines, periosteal irritation, and stimulation of intraosseous nerves. Others postulate that this pain is often secondary to venous engorgement of the tumor, causing increasing mass effect on pain-sensitive nerve roots, dura, and periosteum. Unlike mechanical back pain, biologic back pain is responsive to low-dose steroids. Patients with mechanical pain describe pain exacerbated with movement that is often dependent on the level of spinal involvement. Of particular note, patients with thoracic or thoracolumbar compression fractures will complain of tremendous pain when lying flat, in part because of the extension forces that exacerbate an already unstable kyphosis. Medically, this mechanical pain responds well to narcotics and orthosis therapy and is often exacerbated by activity, standing, and coughing, very similar to the pain experienced in traumatic instability. These hallmarks classically distinguish mechanical pain from biologic pain, which is important for the practicing clinician, because mechanical back pain appears to be more amenable to relief via surgical intervention.


Radiculopathy and myelopathy are late-stage signs and symptoms of the sequelae of neoplastic processes of the spine and most often are secondary to compression from a directly enlarging neoplastic mass, pressure caused by retropulsion of bony fragments, and/or severe kyphosis form vertebral collapse. Thoracic radiculopathies occur in bandlike segmental levels, unlike lumbar, sacral, and cervical involvement, in which pain or weakness usually involves distinct dermatomal distributions. Myelopathy often initially manifests as hyperreflexia, an upward-going plantar reflex and clonus that will often transition to weakness and loss of proprioception, as corticospinal and dorsal column tracts are damaged secondary to compression. Autonomic dysfunction secondary to spinal cord compression or cauda equina compression marks a distinguishing and ominous symptom of advanced cancer disease of the spine, and patients at this stage often come to medical attention with bowel and bladder dysfunction characterized by painless urinary retention or overflow incontinence.




Radiographic Studies and Preoperative Diagnosis


The initial study of choice often selected to evaluate axial spine tumors is plain radiography, which often can localize lesions and is diagnostic in particular cases; however, often this requires a 30% to 60% loss of mineralization before osteolytic lesions can be detectable, making radiography a poor screening test for neoplastic lesions. Remarkably, plain film radiographs can identify kyphosis and scoliosis deformities in weight-bearing patients, unlike magnetic resonance imaging (MRI) and computed tomography (CT) scans, which are done in the supine position; lesions that result in only minimal vertebral collapse may go undetected using MR and CT imaging.


Advances in imaging have dramatically improved the sensitivity for detecting primary and secondary tumors of the spine, and the repertoire of imaging modalities at the disposal of the practicing clinician are constantly undergoing development to help improve diagnosis and resolve tumor burden from surrounding tissue. The most heavily used imaging techniques include MRI, CT, myelogram, positron emission tomography (PET), bone scans, and angiography.


CT and MRI scans have become the standard imaging modalities used to determine the location and extent of tumor burden. MRI provides powerful tissue resolution, and with the use of gadolinium, it can discriminate tumor from surrounding soft tissue. CT scans are often used in conjunction with MRI and are often used to assess the extent of bony destruction and the quality of the surrounding bony architecture. When used in conjunction, CT and MRI can provide enough imaging characteristics that a diagnosis may be made predominantly from these features without further interrogation of the lesion.


Before MRI became widely available, CT myelography was the test of choice for assessing cord compression. This imaging modality, which uses the injection of contrast into the cerebrospinal fluid space, is now often used when patients cannot get an MRI, either because of metallic foreign objects or implanted medical devices. Bone scans, PET imaging, and angiography can collectively provide insightful information to supplement MRI and CT data. Bone scans that rely on a nuclear tracer, such as technetium-99m-methylenediphosphonate, allow for the identification of areas of bone growth or bone breakdown, thus making them highly sensitive, but poorly specific, for identifying neoplastic processes. PET imaging exploits cancer metabolism and the differential uptake of a glucose-labeled radiotracer by cancer cells. The technique has become popular because it allows for whole-body surveillance in a single study, with reasonably accurate anatomic and functional information regarding tumors. Angiography may often be used dually for diagnostic purposes and therapeutic intervention, and it can aid in the diagnosis of primary vascular lesions, such as aneurysmal bone cysts or hemangioma. Alternatively, it may be used to embolize highly vascular pathologies in the spine, such as renal cell carcinoma or melanoma, while at the same time identifying primary segmental feeders to the tumor.


Ultimately, before any surgical intervention and in the absence of a leading diagnosis of the lesion of interest, an accurate biopsy is essential to surgical planning and decision making regarding management. Histopathologic examination is most commonly achieved via CT-guided biopsy. This procedure has become safe, economical, and reasonably reliable; percutaneous CT-guided biopsy of spinal lesions is diagnostically accurate 93% of the time, with higher rates of success associated with high-grade versus low-grade lesions.


Once all the data have been acquired, the decision to manage and treat these patients is complex and relies on a multidisciplinary team that includes spine surgeons, oncologists, and radiation oncologists as well as the patient and family. Compelling arguments for surgery include palliation, decompression, and/or cure with some rare primary lesions; however, all of this is factored in with the context of the patient’s overall survival, prognosis, and extent of disease burden, making the decision to operate a very difficult one given all these variables.




Surgical Decision Making and Patient Expectations


Patients with Primary Osseous Lesions


After a tissue diagnosis is obtained, the decision to operate depends on several factors that include tissue histology and grade, surgical accessibility, patient symptoms, and premorbid conditions. Each case must be evaluated on an individual basis. The histology of some tumors may predict a primarily nonoperative intervention, such as chemotherapy and/or radiation therapy, but patient symptoms or radiographic evidence of instability may dictate a need for operative intervention. For example, patients with multiple myeloma or plasmacytoma may benefit from surgical decompression and stabilization in the setting of acute vertebral collapse with neurologic sequelae.


Primary malignant tumors of the thoracic spine present a very difficult problem for patients and surgeons. En bloc removal of the tumor with wide excisional margins is required for long-term cure in many cases, such as with chordoma or chondrosarcoma histology. The spinal cord, surrounding axial skeletal support, and structures of the mediastinum often make an en bloc resection a formidable task. Although neurologic preservation surgeries have been described, aggressive tumor debulking with focused proton-beam radiation has demonstrated good overall and progression-free survival at 3 years and may be the best option in some situations. In these cases, residual tumor greater than 30 mL, identified on postoperative imaging, had a negative effect on these outcomes.


A thorough discussion regarding the techniques of en bloc resections of primary thoracic spine lesions is beyond the scope of this chapter; however, it should be mentioned that the typical preoperative evaluation will generally not suffice. Preoperative evaluation with the oncology team, a mental health professional, and a physical medicine and rehabilitation specialist will help educate the patient and anticipate postoperative needs. A meticulous operative plan that involves anesthesia and the cardiothoracic, vascular, plastic, and neurologic surgery teams is necessary. In addition, a preoperative consult with the critical care team will help anticipate immediate postoperative needs. Needless to say, before performing an en bloc resection, every detail of the patient’s treatment course should be outlined, with any potential complication anticipated.


Patients with Metastatic Disease


When considering surgery for a patient with metastatic cancer to the spine, a realistic discussion regarding the goals of surgery, an anticipated extended postsurgical course, including rehabilitation, and overall prognosis is paramount. The patient and his or her family must understand that surgery will only be palliative. Reduction in mechanical back pain, preservation or restoration of spinal stability, and neurologic protection are the primary aims of an intervention. The risks of significant operative blood loss and routine spine surgical morbidity, extended recovery, and failure to improve neurologic status must be emphasized with patients and their families. In one retrospective study, 63% of patients recovered or improved their neurologic function at 3-month follow-up. However, it cannot be overstated that a large percentage of neurologically impaired patients may not functionally improve.


Although a recent population-based analysis showed an approximate increase in survival of 50% in patients with breast cancer metastasis to the spine treated with spinal fusion over the past decade, no prospective randomized trial has shown an increased survival rate among surgically treated patients. As with all patients with a terminal illness, the primary goal is to improve quality of life. In a 2005 landmark paper by Patchell and colleagues, patients with metastatic disease and epidural extension with displacement of the spinal cord confirmed radiographically were randomized to surgery with adjuvant radiation or radiation alone. The surgical techniques were not standardized, but statistically significant benefits were found in patients’ retention or recovery of ambulation in the surgery-with-radiation group. Similarly, a recent multicenter observation study demonstrated significantly improved quality of life in those patients who underwent surgical decompression for osseous metastatic disease. This benefit should be stressed, especially in patients who suffer from diseases with relatively long survival times, such as patients with breast, prostate, or renal carcinomas.


In our institution, patients who are medically able to tolerate surgery and have an expected survival of at least 6 months are evaluated for surgery. However, in the setting of an acute neurologic decline, most patients are considered for an intervention regardless of prognosis. Although Patchell and colleagues found that 62% of nonambulatory patients recovered their ability to walk, preoperative status often predicts postoperative status, and this should be considered before surgery. Patients who have complete loss of motor and sensory function before surgery and those who have American Spinal Injury Association (ASIA) class A deficits are rarely considered for surgery.




Preoperative Embolization


Regardless of the surgical approach, thoracic spine tumors have the added morbidity of increased blood loss. The literature is much more robust with retrospective data on benefits in metastatic cases compared with the current practice for primary spinal lesions. Although efficacy has not been established, several authors advocate for preoperative embolization of primary osseous tumors as well. These tumors include chondrosarcoma, chordoma, and osteosarcoma among others. Yang and associates described the benefit derived from preoperative embolization of both primary and metastatic disease to the sacrum in their series of patients.


To date, no prospective randomized trial has been performed to clearly delineate the possible benefits of preoperative embolization for metastatic disease. However, this practice has evolved a large number of single-institution experiences that show benefit. Gellad and colleagues found that preoperative embolization reduced intraoperative blood loss by nearly 50%. Renal cell carcinoma is often hypervascular and is traditionally regarded as the most common metastatic histology with risk of increased intraoperative hemorrhage. Therefore several authors advocate for preoperative embolization of all suspected renal cell carcinoma metastases and have shown a benefit in reduction of blood loss. Schirmer and colleagues found less than anticipated blood loss in patients with renal cell carcinoma metastasis when patients were preoperatively embolized percutaneously with n-butyl cyanoacrylate embol.


Although many surgeons find preoperative embolization to be a useful adjunct for renal cell metastasis, it is not the most useful predictor of blood loss. Rehak and colleagues found an average blood loss of 4750 mL in their embolized group and an average blood loss of 1786 mL in their nonembolized group, a difference they explained largely by the complexity of the surgery. Tumor histology may also play a major role in the benefits of embolization; Robial and associates found a significant decrease in blood loss with renal cell carcinoma but found no benefit in patients with primary breast and lung carcinoma. Although preoperative embolization may not predict intraoperative blood loss, embolization should be considered in any lesion suspected to be hypervascular.




Approaches to the Thoracic Spine ( Fig. 33-1 )


Thoracic Laminectomy


Since its description in the early nineteenth century, laminectomy has become one of the most common neurosurgical procedures performed. Unfortunately, its use in isolation is seldom appropriate for thoracic tumors; rather, it is indicated for use in patients with bony neoplasms that involve the posterior elements of the spine and for tumors with dorsal epidural extension and significant mass effect. As stated, few patients will benefit from laminectomy alone. Before recent advancements in spine surgery, its use was associated with poor outcomes.


Jul 11, 2019 | Posted by in NEUROSURGERY | Comments Off on Surgical Approaches to Thoracic Primary and Secondary Tumors

Full access? Get Clinical Tree

Get Clinical Tree app for offline access