A CT apparatus with a tiltable gantry (usually 20°–30° in both directions) is preferred, while multidetector technology, CT fluoroscopy, and a screen inside the CT room are optional additional and potentially useful features, based on operator’s preference. On the initial volumetric scout views in two orthogonal projections, based on the desired CC angulation of the needle access to the target, the gantry is tilted as necessary, and the localizing scan encompassing the region of interest is performed. The axial slice visualizing the target and an accessible needle path is selected, and the skin is marked at the chosen needle entry site. A radiopaque grid with vertical bars can be applied to the skin and used to define the exact needle entry point on the RL axis. Sliding the CT cradle, the marked skin point is placed under the laser light of the CT gantry, and the needle is aligned from its entry into the skin to its hub with the laser light. If this technical tip is correctly applied, the gantry-needle-target alignment is obtained; the whole needle shaft, from the skin entry point to the tip, and the target are in-plane with the gantry and are visible on one single axial slice; moreover the projected path of the needle is well discernible and predictable (Fig. 3.2). While the needle-gantry alignment controls the CC angulation of the needle, the RL obliquity is left to the operator’s ability, and usually rapidly adjusted after few trials and errors in the superficial tissues, unless some recently available guiding devices are used. It is imperative that the control CT views obtained intermittently during the procedure show the whole length of the inserted needle, with one slice above and one below in which no needle be visualized; if this safety condition is not respected, there is risk for a not perfectly in-plane needle, or for a needle tip slightly curved, to be inadvertently advanced off-plane and out of the operator control and sight, with risk of injury in a complex and delicate anatomical region such as the spine. Strict immobility of the patient is also required between localizing scan and skin marking and also strongly desired during the whole procedure; in fact even small patient’s movements during the procedure may render the CT-guided needle-access complex, lengthy, and in worst-case scenario imprecise.

Note that sterile prepping and draping usually precede skin marking if fluoroscopic guidance is used, while the opposite occurs if CT guidance is favored.

A wide variety of devices, produced by different vendors, each with its own specific features, can be used to obtain spine biopsies (Fig. 3.3). We recommend the use of coaxial systems, composed of an access cannula, which can be a trocar needle, a bone access needle, and a vertebroplasty needle, with diamond or beveled tip, and a biopsy device, to be used coaxially. A beveled tip allows slight steerability of the access cannula during positioning in the vertebra. The coaxial systems allow multiple biopsy passes through the same access, if necessary. The access cannula, of large caliber, can be slightly redirected in a different RL or CC direction, and the coaxial biopsy device inserted along different directions to obtain sampling from a wider region. Coaxial nitinol curved-tip biopsy cannulas exist to obtain multiple samples from different regions of the target lesion, once the access cannula has been placed. The biopsy device can be a cannula with a trephinated or a fish-mouth tip, if an osseous lesion is to be sampled, or a cutting spring-loaded needle, if a soft tissue lesion is being targeted. In case a sclerotic bone has to be traversed, the use of a coaxial drill can be used to create a path for the access cannula; in some cases the use of a surgical hammer is necessary to obtain access-cannula penetration through cortical or sclerotic bone.

In case of osseous lesions with heterogeneous density, it might be advisable to sample the lesion in multiple small increments (5–8 mm) to avoid the risk that a more proximal sclerotic sample in the biopsy cannula crashes the rest of the sample when the cannula is further advanced in the lesion (Fig. 3.4).

We suggest to advance and then to retrieve the biopsy cannula under constant vacuum suction. Vacuum can be easily obtainable with a set of small tubing attached with a Luer lock junction at the cannula and with a three-way stopcock at a large-volume (20–60 mL) syringe. Suction can be held by a second operator, while the first operator handles the biopsy cannula, by locking the syringe in vacuum with a needle holder or by the use of a dedicated self-locking vacuum syringe (Fig. 3.5). Spinal needles of different lengths, usually 22 G in caliber, are used to perform local anesthesia on the periosteum; scalpels are used to make small incision to the skin and if necessary to the fascia, for easier access of the cannulas. In some cases k-wires can be used to adopt the Seldinger technique, in case cannulas of different lengths and/or caliber have to be exchanged without the need to perform a new percutaneous access. We suggest the use of rather large-caliber systems, typically from 11 to 8 G cannulas for bone access, which allows the use of adequate size coaxial biopsy devices to obtain large core biopsy samples (see Fig. 3.3).

Positioning of the patient is individualized based on the planned access, typically in the prone position for access to the sacral, lumbar, thoracic, and posterior elements of the cervical spine, while in the supine position for anterior access to the lower cervical spine and for paramaxillary or trans-oral approaches to the upper cervical spine. Lateral decubitus can be used in selected cases. Bolsters strategically placed under patient’s body can be used to favorably alter spine curves and render certain spine accesses easier (see Fig. 3.6). A typical example is a bolster under the lower abdomen in a prone patient to flatten the lumbosacral lordosis and facilitate access to an otherwise steeply oblique L5–S1 disc space. A general principle is also to favor a decubitus the patient can reasonably hold remaining still and comfortable for the expected duration of the whole procedure, if the biopsy is conducted on an awake patient; for example, a prolonged prone decubitus can be problematic for an obese patient with respiratory problems. Prepping of the skin should be wide and thorough to ensure sterility; when the skin is marked at the entry site with an indelible marker, alcoholic prepping solutions should be avoided, due to the ability of alcohol to promptly delete ink from the skin. We recommend full draping and the use of full gown, gloves, hat, and mask for the operator, in every bone or disc access, due to the heightened risk of infection. Local anesthesia should be delivered to the sensitive tissues, such as the skin, fascia, and periosteum. Local anesthesia close to the nerve roots should be avoided if the needle access has the potential risk of injury to the nerve; for example, no anesthesia should be delivered in the neuroforamen or in the region of the lumbar plexus if a posterolateral direct access to the vertebral body is being performed; the nerves should remain awake and sensitive, so that they serve as a warning if the needle course is too close to them. For the same reason, such accesses should not be attempted in a patient unconscious or under general anesthesia.

General but all relative contraindications to a spine biopsy are an uncooperative patient, unlikely to remain still during the procedure, coagulopathy, low platelet count, and treatment with anticoagulants or antiplatelet agents (spine biopsies are usually performed in patients treated with ASA or other NSAIDs).

The sacrum has a complex tridimensional conformation, and fluoroscopy does not recognize clear and safe landmarks, since the sacral foramina are “V-shaped” channels, and the sacroiliac joints have wavy contours. Fluoroscopy can still be used to guide short- and long-axis needle accesses, especially when a sacroplasty is to be performed, but we recommend CT guidance to biopsy the sacrum, for practicality, precision, and safety (Fig. 3.6). Planning the biopsy, the lesion in the sacrum is viewed in three planes, with multiplanar CT-reconstructed images, and the most suitable access is chosen, ideally along the long axis of the lesion, so that a satisfactory sample can be obtained, keeping the needle path away from the sacral central canal and neuroforamina. When a lytic lesion destroys the cortical margins of the neuroforamen, the needle should strictly avoid the expected course of the nerve root (see Fig. 3.6). In such cases it is advisable to carefully analyze pre-procedural MRI images, commonly able to depict the position of the nerve root even when the foramen has been invaded by a mass. Rather rarely an access through the iliac wing will be necessary to reach a lesion in the sacrum. Trans-sacral approach can also be used when a retroperitoneal presacral mass needs to be biopsied (see Fig. 3.6). Attention should be paid to avoid the nervous structures of the sacral plexus that lie just anterior to the sacral alae. As usual, after bone access, the guiding cannula is placed adjacent to the margins of the lesion, and the biopsy can be performed with a coaxial device, depending on the consistency of the target lesion.

The lumbar vertebrae are characterized by rather large and squared vertebral bodies, pedicles horizontally aligned parallel to the superior end plate, projecting over the superior half of the vertebral body, progressively larger and posterolaterally oblique from L1 to L5 (in fact it should be noted that L1 pedicles can be very thin and straight, so that an oblique trans-pedicular needle access might be impossible), and thin, long, and straight transverse processes. Spatial orientation of the vertebrae, along lordotic and/or scoliotic curves, strongly influences needle accesses.