Preprocedure Evaluation

A complete preprocedure evaluation is an important component of an efficient and effective biopsy service. This evaluation should consist of reviewing prior diagnostic imaging studies, obtaining a bleeding history and appropriate laboratory studies, and obtaining written informed consent. We have found that incorporating physician extenders into the procedural approval process helps ensure that biopsies are performed as efficiently and safely as possible. Integration of physician assistants, nurse practitioners, and nurse coordinators allows the radiologist to focus on the procedure, improving throughput. Furthermore, many of these patient encounters, whether they are performed on an inpatient or elective outpatient basis, can be coded and billed as a specialty consultation, allowing the radiology group to be reimbursed appropriately.

Review of prior diagnostic imaging studies will confirm the presence of a lesion suitable for biopsy and help in planning of a specific approach, choice of the appropriate guidance modality, and characterization of the lesion to provide the pathologist with an appropriate differential diagnosis. The importance of reviewing prior imaging studies cannot be overstated. For example, a radiologist may be asked to biopsy a lesion that, on review, proves to be a benign hemangioma or cyst.

The appropriate laboratory investigation of the patient before a biopsy remains the subject of debate. No single published guideline is widely accepted or used. This lack of consensus stems from the fact that no prospective evaluation of a large number of patients has been performed in which various factors, including patient history, specific type of procedure, and laboratory tests, have been compared with outcome.

Silverman and colleagues proposed a strategy for screening laboratory tests for abdominal interventional procedures based primarily on the bleeding risk of the procedure and screening of patients for bleeding tendencies. Since the work by Silverman and coworkers, consensus guidelines have been published by the Society of Interventional Radiology ( Table 71-1 ). The recommendations are based on available literature and consensus expert opinion but, admittedly, lack high-level evidence. As a result, the recommendations can conflict with other publications. For example, in a review by O’Connor and associates, the international normalized ratio (INR) and platelet transfusion threshold for percutaneous liver biopsy were 2.0 or lower and 25,000/mL or higher, respectively, whereas, the Society of Interventional Radiology guidelines recommend an INR and platelet transfusion threshold of 1.5 or lower and 50,000/mL or higher, respectively. The lack of high-level data contributes to the nonuniform pattern regarding management of hemostatic defects. It is likely that individual practitioners will tailor their guidelines to local expertise and patient comorbidities.

A patient’s bleeding risk is also influenced by medication use, with emphasis on anticoagulant and antiplatelet agents. Anticoagulant medications have traditionally included warfarin and unfractionated heparin. Newer anticoagulants include low-molecular-weight heparin (e.g., enoxaparin), indirect factor Xa inhibitors (e.g., fondaparinux, idraparinux, idrabiotaparinux), direct Xa inhibitors (e.g., rivaroxaban and apixaban), and direct thrombin inhibitors (e.g., lepirudin, argatroban, bivalirudin, dabigatran). The lack of objective data evaluating the periprocedural management of patients receiving anticoagulant or antiplatelet agents makes the proposal of general recommendations for the interventionalist difficult. The American College of Chest Physicians has proposed stratifying patients on the basis of risk for perioperative thromboembolism. Patients are classified as being at low, moderate, or high risk for thromboembolic events ( Table 71-2 ). By use of this classification scheme, patients with the highest thromboembolic risk and scheduled to undergo a procedure with a high hemorrhagic risk (e.g., renal biopsy) stand to benefit the most from interruption of anticoagulation and bridging with an anticoagulant with a short half-life (e.g., enoaparin). In observational studies, this regimen was associated with a 1% to 2% incidence of thromboembolic events in the high-risk group. There is a paucity of evidence to guide the use of bridging anticoagulation for moderate- and low-risk categories.

If anticoagulation can be withheld, it is frequently helpful to allow a time lapse of five half-lives, which corresponds to a residual drug activity of 3% from the initial dose. Whereas making decisions based on the half-life of a drug is reasonable, clearance can be affected by drug-drug interactions, differences in metabolism, and genetic influences. If a procedure requires more urgency, an elevated INR may be reversed immediately by administering fresh frozen plasma. Alternatively, vitamin K can be used to reverse the effects of warfarin. An elevation of partial thromboplastin time induced by heparin may be reversed with protamine, a heparin antagonist. Low-molecular-weight heparin (i.e., enoxaparin) has a half-life of 4.5 to 7 hours, based on anti-Xa activity. In general, most percutaneous interventions in the abdomen and pelvis can be performed after withholding of the therapeutic dose on the morning of the procedure.

Similar to anticoagulants, antiplatelet agents can also increase a patient’s hemorrhagic risk during surgery. Platelet inhibitors include aspirin, thienopyridines (clopidogrel, prasugrel, ticlopidine), and glycoprotein IIb/IIIa inhibitors (e.g., abciximab, eptifibatide, tirofiban). Appropriate management of antiplatelet agents is determined by the indication. Most common indications include secondary prevention of ischemic cardiac events, post–coronary stenting, and secondary prevention of cerebrovascular events. Stopping of these agents should be considered carefully. At our institution, it is common to consult the treating cardiologist before cessation to better understand the risks associated with stopping of the medication against the hemorrhagic risk of the procedure. If a cardiac event occurred within 1 year and the patient is taking aspirin or clopidogrel, we generally perform the biopsy without stopping the medication but inform the patient of the increased hemorrhagic risk. Aspirin, clopidogrel, prasugrel, and ticlopidine irreversibly inhibit platelet function, making the half-life of the drug irrelevant. For each day that one of these agents is withheld, approximately 10% to 14% of the normal platelet function is restored, taking 7 to 10 days for the entire platelet pool to be replenished. On the other hand, dipyridamole, cilostazol, and nonsteroidal anti-inflammatory drugs reversibly inhibit platelet function, and their effects are dependent on the elimination half-life. The Society of Interventional Radiology has published consensus guidelines on appropriate management of anticoagulant and antiplatelet medications ( Table 71-3 ).

Written informed consent should be obtained from each patient. The biopsy procedure should be described to the patient thoughtfully in layman’s terms. Patients should be informed of the risk of bleeding and infection and that biopsy of upper abdominal lesions may result in a pneumothorax and possibly chest tube placement. Patients should be informed that multiple needle passes may be required, the specimen may not be diagnostic, and additional work-up may be necessary. Patients with lesions near bowel are at risk of bowel injury and abscess, although this complication, surprisingly, has only rarely been reported. The preprocedure visit is also an excellent opportunity to assess various factors, such as the patient’s airway, ability to lie in the desired position, and level of anxiety. All these variables play a role in deciding the level of sedation (i.e., moderate sedation, often administered by the radiologist, or a higher level of sedation requiring an anesthesiologist). A detailed home care instruction form is reviewed with each patient before the biopsy that explains which symptoms are to be expected after the biopsy and which symptoms raise the question of a complication. This form provides a list of contact telephone numbers in case a complication occurs.

Choice of Modality for Image Guidance

Numerous modalities are available for performing image-guided percutaneous biopsies: fluoroscopy, US, CT (with or without fluoroscopic capability), and magnetic resonance imaging (MRI). Each of these techniques has strengths and weaknesses as well as specific indications, and they are discussed next.

Computed Tomography

CT is widely used for image guidance in the United States primarily because of equipment availability and user preference. CT has the advantages of very high spatial resolution and lack of imaging “blind spots.” Furthermore, the depiction of intervening structures is superb. Disadvantages include the exposure to ionizing radiation, the lack of direct real-time needle-tip visualization, the difficulty encountered in the biopsy of moving lesions, and the high cost. Although CT is limited to the axial plane, the ability to angle the gantry up to 30 degrees allows some limited flexibility in needle placement, particularly in the cephalocaudal direction. An alternative to angling the gantry is to use the triangulation method, in which three points composed of the lesion (A), the skin overlying the lesion (B), and a point either cranial or caudal to the lesion (C) are selected in the same parasagittal plane. The position of C should be selected such that a line formed between A and C does not transgress any critical structures (see later, routes to avoid). These three points form a right triangle, and by trigonometry, the length and angle of insertion can be calculated ( Fig. 71-1 ).

An 87-year-old man with melanoma.

Parasagittal reformation of an intravenous and oral contrast-enhanced CT of the upper abdomen reveals a right adrenal mass. Because biopsy of the lesion through an axial approach would have transgressed lung, the triangulation method was used. Three points are selected: A, the lesion; B, the skin overlying the lesion; and C, a point either cranial or caudal to the lesion. The position of C was selected such that a line formed between A and C did not transgress any critical structures. These three points form a right triangle, and by trigonometry, the needle length distance and angle of insertion were appropriately calculated. Fine-needle aspiration revealed metastatic melanoma.

CT fluoroscopy is capable of providing six to eight lower resolution and low-milliampere images and near real-time needle-tip visualization. This technique reduces the time advantages of US considerably and improves the targeting of moving lesions. It is particularly useful for procedures involving deep structures, such as retroperitoneal masses, or for procedures involving organs prone to respiratory motion, such as the liver. CT fluoroscopy may use a quick check technique, which is analogous to conventional CT. This technique uses single-section CT fluoroscopic images to check needle location and to confirm appropriate alignment. Continuous CT fluoroscopic images may be obtained in the region of the needle when the needle tip is difficult to localize, such as when it is in an oblique or a transverse plane. This technique is analogous to conventional CT, except reconstruction times are faster and the radiologist may manually position the table. Continuous fluoroscopy denotes the use of continuous fluoroscopic exposure during needle advancement or manipulation. It is wise to use forceps as a needle holder to prevent primary beam irradiation of the radiologist’s hands.

Radiation doses to the patient and radiologist are higher in CT fluoroscopy than in conventional CT; however, observed doses have fallen with the trend toward the quick check technique and modulation of two scanner parameters, which are usually readily displayed: CT dose index and dose-length product. These parameters can be lowered by modifying the longitudinal scan length, number of scans, and tube current–exposure time product (milliampere × second [mAs]).

Solid masses, which are isodense to surrounding organ parenchyma, are difficult to biopsy. Intravenous contrast material may be administered to increase lesion conspicuity. We recommend administering intravenous contrast material after placement of the guide needle or biopsy needle near the lesion, based on anatomic landmarks.

Magnetic Resonance Imaging

MRI has been used sparingly for guiding percutaneous biopsies, although the roadblocks to use of this modality are diminishing. The advantages of MRI include high spatial resolution, very high inherent tissue contrast, lack of ionizing radiation, real-time capability, and virtually unlimited multiplanar imaging planes, which facilitates needle placement for lesions not readily accessible with a traditional axial approach ( Fig. 71-2 ). Disadvantages include the requirement for MR-compatible supplies and monitoring equipment, the considerable time commitment, and the high cost. Many of these disadvantages, however, are significantly reduced or eliminated with the open or dedicated interventional units, which allow placement of the needle while the patient is in the bore of the magnet and use fast imaging sequences that provide near real-time guidance. The use of lower field strength in an open system decreases the signal-to-noise ratio and results in longer acquisition times but may still be sufficient for lesion visualization. The high inherent tissue contrast attainable on noncontrast MRI can be a major advantage; in most practices, this modality is used selectively in patients with lesions that are not well seen on US and CT. This imaging scenario, however, is infrequent in the abdomen.

MRI of a 76-year-old woman with ampullary carcinoma.

A. Axial T2-weighted MRI demonstrates a 1.2-cm T2 hyperintense lesion in the hepatic dome ( arrow ). B. Coronal MRI demonstrates the biopsy needle within the hepatic dome lesion ( arrowhead ). Given its multiplanar capabilities, MRI facilitates biopsy of lesions that would be difficult to target by conventional axial approaches. Fine-needle aspiration revealed hepatocytes with focal chronic inflammatory cells.

Choice of Needles

In choosing a needle for image-guided biopsy, the first issue to address is what technique will be used to acquire the sample. A single-needle technique uses a new needle for each pass. This is limited by the necessity of imaging guidance for each pass, resulting in long procedure times, need to traverse structures with each pass, increase in the risk of complications, and increased radiation exposure when CT guidance is used. In the tandem technique, a small-caliber needle is first used to localize the lesion with image guidance. A larger caliber biopsy needle is then advanced parallel to the localizing needle without imaging guidance. This technique is limited by multiple organ punctures and imprecise needle-tip localization. At our institution, most operators use a coaxial technique, during which a guide needle is advanced down to the lesion under imaging guidance. Biopsy needles are then advanced coaxially through the guide needle. The drawbacks include having to use a larger caliber guide needle to accommodate the biopsy needle and that subsequent passes may follow the same path and yield little diagnostic tissue.

Guide needles are available in a wide range of size, length, and tip configuration. In general, the needles used for biopsies in the abdomen and pelvis range in size from 16- to 19-gauge and 5 to 20 cm in length. The tips of the guide needles may have an angled bevel or a stylet with a sharp point. A drawback of the beveled needles is that they may deflect away from the intended target as they pass through tissue interfaces, which renders accurate needle placement somewhat more difficult. Needles with a pointed stylet tend to track along a straight line. The Hawkins-Akins needle (Cook Medical, Inc., Bloomington, Ind) also contains an interchangeable blunt stylet, which reduces the risk of injury to bowel, nerves, and blood vessels.

Many biopsy needles are available. These can be broadly grouped into aspirating and cutting needles ( Fig. 71-3 ). The aspirating needles are usually 20- to 25-gauge and are designed to yield individual cells or small clumps of cells that can be spread into a single cell layer for cytopathologic analysis. The “skinny” 22- or 25-gauge needle, although in widespread use, is very flexible and particularly susceptible to bending and deflection. At times, however, the purposeful bending of such a thin-gauge needle can aid in targeting lesions that would be difficult to access otherwise. A curved needle placed coaxially through a straight needle is advantageous because it can compensate for inaccurate guide needle placement and can also sample different regions of a lesion without having to manipulate the outer needle. If a curved needle is used without a guide needle, care should be taken in inserting the needle as rotation of the needle may result in a lacerating effect.

Biopsy needles.

Many radiologists attach a syringe and tube to the needle to apply suction during the actual biopsy. We have abandoned the use of this suction method in favor of simply removing the stylet and relying on natural capillary forces and mechanical agitation to draw tissue into the needle. The main advantage of the nonsuction technique is that the specimens are usually free of blood. Fibrin clots form quickly within bloody aspirates, rendering them difficult to smear onto a glass slide. Also, the presence of abundant erythrocytes obscures cellular detail.

The cutting needles, usually 14- to 20-gauge, are designed to obtain a core of tissue suitable for histologic analysis. Most radiologists have adopted the use of automated cutting needles. These automated needles have an inner slotted stylet for the specimen and an outer cutting stylet. They consistently provide an excellent core of tissue. Manufacturers have designed single-use automated or semiautomated cutting needles that are so lightweight they will maintain their position during the movement of patients in and out of the CT gantry. Cutting needles with a short, long, or adjustable excursion are available. Many of these needles lend themselves to a coaxial technique, permitting several biopsy samples to be obtained from a single skin and organ puncture. In a blinded evaluation of 20 automated cutting biopsy devices, the best overall performance was obtained with 18-gauge needles with at least a 2-cm excursion.

It has been suggested that radiologists should use the smallest gauge needle possible in performing biopsy procedures. Researchers have explored the effect of needle gauge on organ bleeding in the pig model. This work shows that in general, larger needles produce greater bleeding. The research also shows that large needles yield greater amounts of tissue. To the extent that each needle pass carries risk, the maximum tissue yield can be obtained at minimum risk by performing fewer passes with a larger needle. There are two caveats to consider. First, cytopathologists prefer to analyze a thin layer of single cells or clumps of cells. Samples obtained from thin needles (i.e., 20- to 25-gauge) may be easier to smear into a single cell layer than samples from larger needles (14- or 18-gauge). Second, use of cutting needles is riskier than use of aspirating needles. If the knifelike blade of the cutting needle encounters an artery or a vein, the vessel will be lacerated and bleed. In contrast, aspirating needles tend to displace rather than to cut tissue.

MRI is used to guide tissue biopsies particularly in the central nervous system and breast. Dedicated MR-specific needles are now available, although the selection of biopsy needles and sizes is considerably more limited than with the ferromagnetic needles traditionally used in US and CT cases. These nonferromagnetic needles are readily visualized as a signal void and are safe to use in the magnetic field. Use of a ferromagnetic needle, on the other hand, can cause considerable image distortion that may obscure the lesion of interest and hinder precise needle localization. In addition, they may be torqued or deflected in the magnetic field, raising questions about their safety.

It is vital to coordinate needle selection with the pathologist who will interpret the case. If the pathologist is skilled in cytopathology, small-bore (20- to 25-gauge) aspirating needles are recommended. If the pathologist prefers samples for histologic analysis, larger bore cutting needles are appropriate. Some groups perform a cytopathologic touch preparation for samples obtained with core needles. This technique allows a rapid preliminary diagnosis and preserves the core material for permanent fixation and sectioning.