Imaging Techniques

Several factors are considered in selecting the most appropriate imaging modality to guide percutaneous tissue sampling. These include the site, size, and depth of the lesion; its proximity to the pleura and neurovascular structures; and performance status of the patient.

Integrated Positron Emission Tomography–Computed Tomography

Positron emission tomography (PET) has an established role in the detection and staging of neoplastic diseases. In lung cancer it provides information on the primary lesion, early nodal involvement, and distant metastases. The positron-emitting agent most frequently used is ¹⁸F-fluorodeoxyglucose (FDG). This tracer accumulates at sites of increased glycolysis (e.g., tumor cells), and this activity is then detected by the PET camera. The intensity of activity is displayed on a color scale, and a quantitative assessment is made by measuring the standardized uptake value (SUV).

CT combined with PET allows accurate anatomic localization of FDG-avid foci. This technique is particularly useful in isolating active foci surrounded by benign changes, such as within thickened pleura, or in separating tumor from collapse, allowing greater precision in positioning the biopsy needle (Figure 13-3).

Figure 13-3 Differentiating central tumor from distal lung collapse. A, The computed tomography (CT) scan shows central low-density tumor in the left lower lobe (star) with distal collapse (arrowheads), but differentiation is subtle. A small pleural effusion (arrow) is evident. B, CT–positron emission tomography (PET) study demarcates the FDG-avid tumor from the area of distal collapse. Note normal physiologic left ventricular avidity. FDG, ¹⁸F-2-fluorodeoxyglucose.

The sensitivity of PET is limited by the size of the lesion. Small lesions (usually less than 1 cm) may not accumulate sufficient FDG to be detected on PET imaging, leading to false-negative results. The latter can also occur in tumors with relatively low metabolic activity such as carcinoids and alveolar cell cancers.

Inflammatory conditions such as bacterial pneumonias, abscesses, tuberculosis, and active sarcoidosis are associated with increased granulocytic activity. Such activity promotes increased uptake of FDG, potentially giving rise to false-positive results.

Needles

The two main modes of sampling are fine needle aspiration (FNA) for cytologic study and core biopsy for histopathologic examination. Both methods retrieve samples that are suitable for culture. Although the sensitivity of FNA is improved by having a cytologist present to ensure that an adequate sample is obtained, the diagnostic yield in benign disease remains low (20% to 50%) compared with that for core biopsy (70%) (Greif et al., 1999). In malignant disease, the techniques are analogous, with a sensitivity of 90% to 95% (Klein et al., 1996). The clinical requirement for cores of tissue for immunohistochemical analysis has led to a reduction in the use of FNA, and when feasible, cores are always obtained.

FNA involves inserting a fine needle into a lesion and carefully moving the tip to and fro within the tissue, to obtain an aspirate. Gentle suction can be applied by attaching a 5-mL syringe to the needle hub. A variety of fine needles are available, including the Westcott, Chiba, Franseen, and Rotex needles, which range in size from 20 to 23 gauge. Larger-gauge needles are smaller in caliber and less rigid in design, which can limit their use. The Westcott needle has the advantage of having a trough close to the needle tip, which captures small cores of tissue in approximately 50% of cases. The main use of FNA is to obtain nodal aspirates.

Core biopsy specimens are obtained using cutting needles, which are larger in diameter (14 to 18 gauge) and are available in different lengths (6-, 9-, and 15-cm lengths are commonly used). These needles are more sturdy, allowing greater control in placement. Cutting needles are typically mounted on a spring-loaded mechanism. Older designs such as the Bard Biopty biopsy system, when triggered, simultaneously fire an inner notched stylet and an outer cutting cannula. The handle of the Bard system is bulky but reusable (Figure 13-4). Newer, lighter designs include the Cook Quickcore and Bauer Temno devices, which allow the inner notched stylet to be advanced and secured within the lesion before the cutting cannula is activated. The throw can be increased from 1 to 2 cm to obtain better core samples (Figure 13-5). The Cook device seems to have a slightly sharper needle tip, which in practice can be advanced through tougher tissues.

Figure 13-4 The Bard Biopty biopsy system. The needle is introduced at a point on the thorax so that the tip reaches the area in which biopsy will be performed.

Figure 13-5 The Bauer Temno biopsy instrument. The cutting needle is positioned within the lesion to be biopsied by pushing the plunger. The advantages include that the instrument is lightweight and requires use of only one hand, and that it is easy to use under computed tomography or ultrasound guidance.

Historically, it was believed that use of larger cutting needles carried higher complication rates than those associated with fine needles. A 2002 study of practice based in the United Kingdom analyzed data from 5444 lung biopsy and FNA procedures and found no difference between the two methods, a conclusion supported by other studies (Richardson et al., 2002).

With transpulmonary biopsy, the risk of pneumothorax is more closely related to the number of pleural passes. The introduction of a coaxial system has transformed lung biopsies. A single pleural pass is made with a thin-walled introducer needle (usually 16 gauge). A smaller cutting needle (usually 18 gauge with a 1- or 2-cm throw) is inserted through the introducer and multiple cores are taken without repuncturing the pleura. The process is quicker, simpler, and safer than attempting multiple pleural passes, leading to a reduction in the complication rate with an improved diagnostic yield (Figure 13-6).

Figure 13-6 The Cook coaxial system. A 16-gauge introducer needle, together with its stylet and 18-gauge inner cutting needle, is shown.

Indications

The role of percutaneous lung biopsy in diagnosing malignant disease is well established, with a sensitivity of 90% to 95%. Its main application is in patients with inoperable lung cancer, when sputum cytology and bronchoscopy are nondiagnostic, to provide a means of establishing cell type before chemotherapy and radiotherapy. In the past, both FNA and core biopsy have been comparable in diagnosing and distinguishing small cell lung cancer (SCLC) from non-SCLC (NSCLC), providing oncologists with sufficient information to make choices regarding appropriate chemotherapy regimens. New targeted chemotherapy and biologic agents (e.g., the epidermal growth factor receptor [EGFR] inhibitor drugs) have proven prognostic benefit with particular histologic subtypes. This has meant that core sampling is essential to allow routine performance of detailed tissue analysis. FNA can be adequate to differentiate between subtypes of NCSLC, but it remains less accurate overall than core biopsy. The latter allows sufficient material to be obtained for both accurate histologic subtyping and molecular testing (e.g., to determine EGFR mutation status) (Barnes et al., 2010).

Biopsy also is widely used to exclude lung metastases in patients with potentially operable lung tumors, and in those with extrathoracic malignancies (Figure 13-7).

Figure 13-7 Computed tomography (CT)-guided biopsy of a pulmonary nodule using a coaxial system. The patient is in the prone position. Multiple cystic and solid pulmonary nodules in a patient with a history of lymphoma. The diagnosis was Langerhans cell histiocytosis.

With solitary pulmonary lesions, core biopsy remains the prudent approach for confirming benign disease. The combined use of CT guidance and the coaxial biopsy system, which allows multiple cores to be taken safely, has improved the diagnostic accuracy of this technique in both benign and malignant disease. If malignancy is strongly suspected, it may be best to avoid biopsy and proceed straight to surgery.

Transthoracic biopsy and FNA also have a role in the diagnosis of non-neoplastic disease. Both techniques are increasingly being used to obtain samples for identification of microorganisms, particularly in immunocompromised patients with consolidation and masses. All samples are routinely sent to both histology (or cytology) and microbiology. Communication with the referring clinician is essential. In cases in which infection is suspected, the first sample should be sent to microbiology should the procedure need to be unexpectedly abandoned. The converse is true in suspected cases of malignancy where the first sample is sent to histology. The working diagnosis may also influence the choice of needle size.

Contraindications

The contraindications to percutaneous lung biopsy are largely relative and are summarized in Table 13-1. In general, patients need to be able to cooperate, including lying still in the desired position, resisting excessive coughing, and controlling breathing. In patients with very poor lung function, biopsy of a lesion is still possible, provided that the lesion is peripheral and a carefully considered route is identified (usually with CT) that does not traverse lung parenchyma. In performing the biopsy, care is taken to avoid creating a pneumothorax, which could be life-threatening. Before the procedure, a coagulation screen is performed according to local protocol, and bleeding diatheses are corrected, when appropriate, to keep within safe guidelines. Biopsy of parenchymal hydatid lesions has been documented, but an increased and probably unacceptable risk of anaphylactic reaction has been documented in patients with such lesions.

Table 13-1 Contraindications to Needle Biopsy

FEV₁, forced expiratory volume in 1 second.

Method

Pulmonary lesions are usually sampled under CT guidance using cutting needles rather than fine needles. The coaxial technique is preferred because it allows multiple consecutive cores to be obtained without delay, helping to improve diagnostic accuracy while minimizing risk. The number of biopsy specimens varies, but in practice, two to five samples are required (fewer if the patient is at high risk of complications). The aim is to obtain a diagnosis without compromising patient safety.

Peripheral lesions located away from fissures and large vessels are ideally targeted, to minimize complications of pneumothorax and hemorrhage. When possible, the approach should cover the shortest distance from the skin surface to the lesion, using a route that allows the patient to be positioned comfortably for a period of at least 30 minutes. The latter consideration is important, because patients are often elderly or frail and breathless at rest.

In lesions obscured by atelectasis, intravenous contrast can be administered to localize abnormal tissue from collapsed lung. PET-CT is useful in identifying avid foci in these cases. If the lesion is cavitating or exhibits central necrosis, the biopsy must be taken from the periphery of the lesion.

Procedure

Practical Aspects

The patient is positioned supine or prone, depending on the anteroposterior location of the lesion. Pillows are used to elevate one side of the chest if necessary, and the arms are positioned to optimize access, avoiding elevation above the head, which typically causes shoulder discomfort in elderly persons.

Radiopaque surface skin markers are placed 0.5 to 1.0 cm apart over the region of interest, and a limited CT scan covering the area is performed with the patient in gentle respiration (Figure 13-8). Once the target lesion is identified, the safest percutaneous route to the lesion is mapped along the axial and sagittal axes, using the CT slice number and surface markers, respectively, as a guide. A cross is drawn on the patient’s skin to mark the corresponding entry site and a sterile field is created. Lidocaine 1% (10 to 15 mL) or 2% (5 mL) is infiltrated down to the pleura, with care taken not to puncture the visceral pleura or lung. The patient is warned that it is not always possible to anesthetize the pleura fully, because of the risk of pneumothorax from the anesthetic needle.

Figure 13-8 Computed tomography (CT)-guided biopsy of a right lower lobe nodule (arrow). Surface grid markers, together with slice position, are used to determine the site of needle insertion.

With use of the coaxial system, the introducer needle is advanced first, through a small skin incision. Because formal breathing instructions can be confusing to the patient, the respiratory phase is judged from movements of the chest wall. A slight “give” is felt when the pleura is breached, and this thrust should be performed with a smooth, firm motion to avoid shearing the pleura, thereby increasing the chance of a pneumothorax. The progression of the needle toward the lesion and the final position of its tip are intermittently checked by repeating the initial scan as required. One-centimeter markers visable along the length of the introducer needle are a useful guide when advancing the needle. Accurate placement of the needle tip is essential and may take time with small and mobile lesions.

Once the introducer needle is optimally positioned, the central stylet is removed and a finger placed over the hub to prevent air embolism. The cutting needle is then inserted through the lumen of the introducer, and multiple cores are taken, with replacement of the stylet between biopsy maneuvers. The tip of the cutting needle can be angled differently with each thrust, to obtain a representative sample—a technique that is particularly useful when tumors show histologic heterogeneity. Saline (2 to 3 mL) can be injected and aspirated through the introducer needle to obtain microorganisms, when indicated.

Once sufficient samples are obtained safely, the introducer needle is removed in a single swift motion during expiration, and a postbiopsy scan is performed if a pneumothorax is suspected.

Complications

A risk-benefit analysis should always be made before any procedure, particularly in high-risk groups (Table 13-2).

Table 13-2 Complications of Needle Biopsy

The most common complication after lung biopsy is pneumothorax (Figure 13-9). Most pneumothoraces are small (less than 2 cm on chest film) and are managed conservatively. Those requiring intervention usually are detected early: 88% immediately after biopsy and the remainder after 1 hour. A delayed presentation is rare. Intervention can involve aspiration by way of a three-way tap or drainage through a one-way (Heimlich) valve; drainage is performed at a frequency of 0 to 17%. Positioning the patient with the puncture site dependent can reduce the pneumothorax rate and should be considered in patients at high risk. Surprisingly, use of a cutting needle does not increase the incidence of pneumothorax over that with fine needle aspiration, the pneumothorax rate being in the range of 3% to 42%. Tension pneumothorax is rare but may be seen in patients with emphysema. It develops within minutes and constitutes a medical emergency.

Figure 13-9 A, Coaxial computed tomography (CT)-guided biopsy of a solitary pulmonary nodule in the left lower lobe in a patient with a history of colonic carcinoma. The needle is angled away from the aorta (Ao). B, After biopsy, a tiny pneumothorax (arrow) can be seen, along with the needle tract and perilesional hemorrhage (arrowheads). The diagnosis was carcinoid.

Minor perilesional hemorrhage is common (see Figure 13-9); however, actual hemoptysis occurs in only 4% to 5% of biopsy procedures (Richardson et al., 2002) and is more common in patients with pulmonary hypertension. It often follows a bout of coughing and can be extremely frightening for the patient. Supportive measures are initiated with the patient positioned biopsy side down. Treatment is rarely required, because spontaneous recovery is rapid. Massive hemoptysis rarely occurs and can be fatal.

Although rare, a systemic arterial air embolism is a serious and increasingly recognized complication of interventional lung procedures, and its effects can be clinically subtle (Hiraki et al., 2007). It occurs when air enters into the pulmonary venous circulation, is pumped through the left side of the heart, and subsequently enters the cerebral and coronary arteries, which become occluded. Air may enter the pulmonary veins by two main processes. It can be drawn in from the atmosphere, through the introducer needle on deep inspiration, when the hub is temporarily exposed in between biopsies. Entry of air through this route is avoided by promptly covering the hub with the thumb each time the cutting needle or stylet is withdrawn, and by asking the patient to hold the breath during the process. Alternatively, air can enter the pulmonary veins through the airways along the needle tract during a bout of coughing. To prevent this from occurring, needles should be removed during uncontrollable coughing. Rapid diagnosis and treatment of a pulmonary venous air embolism are essential. The patient should be placed in the right lateral decubitus position, and 100% oxygen should be administered while awaiting transfer to a hyperbaric oxygen chamber if available. A CT scan of the head and chest should be performed to confirm the diagnosis.

Reported mortality rates for transpulmonary biopsy range from 0.07% to 0.15% (Richardson et al., 2002; Tomiyama et al., 2006). Death may result from cardiac arrest, systemic arterial air embolism, tension pneumothorax, or hemorrhage.

Pitfalls and Controversies

In malignant disease, percutaneous sampling of pulmonary lesions carries a high sensitivity with use of the coaxial system, even if the nodules are small. This has reduced the need to recall patients for repeat biopsies, because the sensitivity for detection of malignancy has risen to 90% to 95%.

In benign disease, FNA has a relatively poor diagnostic yield. Sensitivity is improved by obtaining multiple cores using a cutting needle and the coaxial system, a technique that has been used in the diagnosis of cryptogenic organizing pneumonia, Wegener granulomatosis, and some infections.

Diagnostic difficulties still arise with lesions exhibiting a high level of fibrosis—for example, metastatic breast carcinoma and Hodgkin disease. Excess mucus production as in bronchoalveolar carcinoma also can interfere with establishing the diagnosis.

Biopsy of a solitary mass with suspected malignancy that is potentially operable remains controversial, because the lesion is likely to need removal. In any case, the advantages of biopsy are that the patient and the surgeon are better informed, and that the operation time is shortened by avoiding the need for frozen sections.

There remain four indications for biopsy of a noncalcified solitary pulmonary nodule: First, if benign disease is suspected, biopsy may obviate the need for surgery; second, in a lung cancer patient who is a poor operative candidate, it may be necessary to obtain tissue to optimize treatment with nonsurgical therapies (radiotherapy with curative intent); third, in a patient with an extrathoracic malignancy where a metastasis is suspected; fourth, a fully informed patient may wish to confirm the diagnosis of cancer, even if the suspicion of cancer is high, before proceeding to surgery.

Preparation

Ultrasound-guided sampling of superficial lesions usually requires no fasting or formal hospital admission, and observations are recorded only if the patient history includes clinical concerns. Hematologic indices are checked and optimized beforehand, and antiplatelet therapy is withheld, in accordance with protocols for CT-guided procedures.

Practical Aspects

The site of the lesion or tissue usually determines the optimal positioning of the patient. For example, nontargeted pleural biopsy requires that the patient sit upright, facing away from the radiologist, with the arms folded to spread apart the shoulder blades. Supraclavicular node sampling, on the other hand, is performed with the patient supine, with the shoulders raised on a pillow, allowing the neck to be slightly hyperextended. Time spent ensuring that both patient and radiologist are comfortable helps ensure a quick procedure, avoiding unwanted “shuffling.”

With targeted biopsy procedures, the ultrasound probe (curvilinear or linear) is adjusted until the lesion can be seen along its greatest length. Once this is achieved, the probe is held secure, and a line is drawn on the patient’s skin to mark the orientation of the probe, which is then removed. The skin entry site is prepared, and a cryogesic (ethyl chloride) freezing spray can be applied to numb the skin, particularly in young patients.

Under sterile conditions, the probe is reapplied to the skin (using the marker as a guide to positioning), and local anesthetic (up to 10 mL of 1% lidocaine) is infiltrated down to the lesion using a 21 gauge (green) needle. Deep to the skin, the needle is kept in view by advancing it directly in line with the transducer. The angle between the needle and the skin is adjusted depending on the depth of the lesion. Administering local anesthetic reduces discomfort for the patient, creates a passage to the lesion, and also gives the radiologist an idea of how to angle the larger biopsy needle.

Depending on the size of the lesion, a cutting needle with a 1- or 2-cm throw is selected. Adjacent lymph nodes can be lined up and sampled simultaneously, enabling the 2-cm throw to be used more often, which is preferable, because it provides a better specimen (Figure 13-10). After a small skin incision is made, the biopsy needle is introduced and carefully observed as it is advanced toward and into the lesion. With the tip of the needle secured well within the lesion (which often requires a firm nudge to pierce its capsule or wall), the whole apparatus can be flattened out, so that the inner stylet can be advanced through but not out of the lesion. Once the stylet is optimally positioned, the biopsy specimen is taken by pushing the trigger, after the patient is warned to expect a clicking sound. The needle is then swiftly withdrawn, and pressure is applied to the skin to control any bleeding; the process is repeated as appropriate after the first sample is retrieved.

Figure 13-10 A, Coronal computed tomography (CT) image showing a large right paratracheal lymph node mass (star) and right supraclavicular lymphadenopathy (arrow). B, Ultrasound-guided biopsy of the supraclavicular lymphadenopathy in the same patient. Two adjacent nodes have been targeted, allowing for a longer, 2-cm sample to be taken. The recess of the biopsy needle is seen (double arrow) before triggering of the cutting mechanism. The diagnosis was small cell lung cancer.

Aftercare

After an uncomplicated superficial biopsy procedure, the patient can be discharged home after a brief period of observation, if clinical stability is maintained. Advice is given on necessary precautions and signs and symptoms necessitating urgent medical attention. High-risk patient groups (e.g., elderly persons and those with an underlying coagulopathy) and high-risk procedures (e.g., targeted liver biopsy, as discussed later on), are the exceptions and require closer observation or an overnight stay.

Method

Depending on the location, mediastinal lesions can be accessed endoscopically (by the transbronchial or transesophageal route), surgically with VATS or through an anterior mediastinotomy, or percutaneously under image guidance, if a safe route that avoids traversing the lung is available.

CT is preferred to ultrasound imaging to guide percutaneous mediastinal biopsy, because it helps to map a direct route through the chest wall to the mass, which avoids puncturing lung or vessels (particularly the internal mammary vessels). Contrast-enhanced CT and color Doppler ultrasound techniques are essential tools for assessing vascularity.

With large mediastinal masses abutting the chest wall, a biopsy route that avoids the lung parenchyma is readily available (Figure 13-11). With smaller lesions, a safer route may be achieved with use of a transpleural approach through an existing effusion or pneumothorax. Saline can be injected into the pleural space where there is no effusion, or directly into the mediastinum, providing an extrapleural route. Placing the patient in the lateral decubitus position to shift the mediastinum also can help to create an extrapulmonary route. These alternative approaches are associated with reduced incidence of complications.

Figure 13-11 A, Contrast-enhanced computed tomography (CT) scan shows a large anterior mediastinal mass with central low-density necrosis (star). B, Integrated CT–positron emission tomography (PET) study shows FDG avidity within the mass, with a central photopenic area corresponding to the necrosis on CT scan. C, CT=guided biopsy. The approach is medial to the left internal mammary artery and targets the non-necrotic portion of the tumor. AAo, ascending aorta; DAo, descending aorta; FDG, ¹⁸F-fluorodeoxyglucose; IMA, internal mammary artery; LPA, left main pulmonary artery.

Core biopsy specimens are necessary to diagnose primary mediastinal tumors (e.g., lymphomas, thymomas). Multiple cores for immunohistochemistry studies are required to differentiate lymphoma from other lesions and can be obtained using the coaxial needle technique.

Limitations and Complications

Breathing can cause small lesions to move out of the biopsy plane and, with use of the extrapleural route, can result in inadvertent introduction of the biopsy needle into the lung or pleura. With biopsy of anterior mediastinal lesions, damage to the internal mammary vessels can result in serious hemorrhage. When feasible, a parasternal approach is adopted, with use of CT guidance to minimize this risk. With biopsy of posterior mediastinal lesions, a paravertebral route can occasionally cause injury to the paravertebral and intercostal neurovascular structures, and the azygous vein.

Method

Mediastinal drainage is performed under CT guidance using a Seldinger technique. After administration of a local anesthetic, a 17 or 18 gauge, long puncture needle is advanced into the collection, often under gentle suction applied with a syringe. This technique allows aspiration of the contents when the fluid collection is breached and also helps to confirm the position of the needle tip. The syringe is removed, and a guidewire is fed through the needle into the fluid collection. With the guidewire held firmly at the entry site, the needle is withdrawn and a tract is dilated over the guidewire. The drain then replaces the dilator, and the guidewire is withdrawn. Usually size 10 or 12 French (10F or 12F) drains are used, depending on the viscosity of the fluid.

Method for Non-targeted Pleural Biopsy

If no discrete pleural mass is present, or there is simply diffuse pleural thickening, a non-targeted ultrasound-guided biopsy is performed. This procedure can be undertaken only if sufficient pleural fluid is present. A suitable biopsy site is identified on ultrasound imaging, and the depth from the skin to the parietal pleura is measured. Local anesthetic is administered down to the pleural space until fluid is aspirated. The depth of the anesthetic needle at this point also gives an estimate of the distance from the skin to the parietal pleura. Guided by these measurements, the tip of the biopsy needle is positioned proximal to the parietal pleura, so that when the stylet is advanced, the notch traverses the width of the pleura to enter the pleural space. The core obtained should therefore contain the full thickness of the pleura and some pleural fluid.

Method for Targeted Pleural and Chest Wall Biopsies

Targeted pleural and chest wall biopsies are indicated to sample a discrete mass and are usually performed under ultrasound guidance (Figure 13-12). If visualization is difficult with use of ultrasound imaging, or if the lesion is small, with insufficient pleural fluid, biopsy is performed under CT guidance. The optimum route is one that runs along the main axis of the pathologic process (i.e., an oblique tract), allowing more of the lesion to be sampled with less risk of pneumothorax (Figure 13-13).

Figure 13-12 A, Contrast-enhanced computed tomography (CT) scan shows a heterogeneous pleural mass (arrows) that is invading the chest wall and destroying the adjacent rib (arrowhead). A small pleural effusion is evident. B, CT–positron emission tomography (PET) study was performed for staging and shows that the mass is FDG-avid. C, Ultrasound image demonstrates the mass (cursor at 1, 2). D, Ultrasound-guided biopsy, with the recess of the needle advanced through the tumor (double arrow). The diagnosis was metastatic adenocarcinoma. FDG, ¹⁸F-fluorodeoxyglucose.

Figure 13-13 Computed tomography (CT)-guided biopsy of a rib metastasis (arrows). The diagnosis was non–small cell carcinoma.

Several cores (at least three) are taken, and if the pleura is not particularly thickened, multiple passes may be required. Radiologists favor the 18 gauge Cook Quickcore or Temno cutting device. The coaxial system can be used for pleural biopsy procedures performed under CT guidance. Samples are sent for microbiologic and histologic examination. If the pleural lesion is large and the patient remains asymptomatic after biopsy, no chest radiograph is required before discharge.

Diagnostic Yield

Imaging-guided pleural biopsy using a cutting needle has a sensitivity of 88%, specificity of 100%, and an overall accuracy of 91% for detection of malignant disease.

Method

Hepatic lesions usually are biopsied under ultrasound guidance, where they are easily imaged and rapidly sampled (Figure 13-15). Owing to the risk of bleeding and the potential need for intervention, preoperative preparation of the patient includes fasting for 6 hours before the procedure.

Figure 13-15 Ultrasound image showing a hypoechoic liver metastasis (arrows), which can be targeted for biopsy.

Ultrasound scanning is performed with the patient supine, with the right arm elevated to allow access to the right upper quadrant. The liver is scanned with a curvilinear probe, and the optimum target is identified. Metastatic deposits can be well- or ill-defined and will vary in size, number, echogenicity, and echotexture, but most often they are solid and hypoechoic relative to the surrounding parenchyma. The lesion that can be reached by the shortest intraparenchymal route while avoiding the gallbladder and the hepatic and portal vessels (which are highlighted on color Doppler) is selected for sampling.

With strict adherence to aseptic technique, local anesthetic is infiltrated down to the liver capsule, which has a rich neurovascular supply. The tip of the anesthetic needle is kept in full view to avoid breaching the capsule, because this can increase the risk of bleeding. A small skin incision is then made, and a larger cutting needle is introduced (usually 18 gauge); radiologists at our institution favor the traditional Bard Biopty biopsy system, which obtains a 1-cm core. The biopsy sample is taken with the patient in arrested respiration to limit movement. If insufficient tissue is obtained with a single core, the biopsy can be repeated. However, careful risk-benefit analysis should be made, owing to the increased risk of hemorrhage associated with multiple passes (Grant et al., 1999). In practice, no more than two cores are usually taken. Although multiple cores are avoided, the coaxial system can still be useful, as hemostatic agents (such as Gelfoam sponge) can be inserted down the introducer needle as it is withdrawn, helping to control bleeding along the biopsy tract.

Aftercare

Post biopsy, patients are observed for at least 6 hours. Blood pressure and pulse are recorded at gradually increasing intervals to observe for signs of early bleeding, and analgesia is prescribed if pain develops. Patients can remain supine or lie right side down, with no evidence to support a particular position.

A study of more than 9000 liver biopsies showed that the risk of hemorrhage not only is related to the number of passes made but also is closely associated with the presence of malignancy and the age of the patient (McGill et al., 1990). Given these risk factors, in the context of malignant liver disease, patients undergoing targeted biopsy usually are admitted overnight for careful observation.

A retrospective study of 68,276 percutaneous liver biopsy procedures (performed over a decade) demonstrated that the majority of complications occur early on (61% in the first 2 hours), with most detected within a day (96% within 24 hours). Death was uncommon (with 9 deaths reported per 100,000 procedures) but occurred in patients with either underlying malignant disease or cirrhosis and was due to intraperitoneal hemorrhage, with signs of bleeding seen within 6 hours (Piccininio et al., 1986). After 24 hours, if clinical status remains stable and pain has subsided, the patient can be discharged home with advice to return if any delayed symptoms are experienced.

Indications and Contraindications

Common causes of exudative effusions are malignancy and empyemas. In patients with a community-acquired pneumonia, 40% develop a parapneumonic effusion. Most cases resolve spontaneously, but the effusion can progress to form an empyema. Early and rapid drainage of an empyema is essential to prevent fibrin deposition, which potentially leads to pleural fibrosis and possibly a restrictive defect. An esophageal rupture or endobronchial lesion should be considered in a patient with an empyema without an obvious predisposing cause.

A pleural effusion can be tapped for diagnostic purposes, and a pleural biopsy performed at the same time if indicated. A pleural effusion can be drained for therapeutic purposes if it is large and the patient is symptomatic (usually with a malignant effusion), if it is infected or is an empyema (if the glucose is low, or the pH is less than 7.2), or if a hemothorax is present (usually after trauma). A bleeding disorder is a relative contraindication to pleural drainage, predisposing to hematoma formation.

Technique

Ultrasound imaging is the easiest and quickest way to confirm the presence and volume of pleural fluid and also demonstrates the presence of echogenic material, septa, and multiple locules, which may require multiple drainage tubes (Figure 13-18).

Figure 13-18 Ultrasound image showing a complex pleural effusion with multiple septations.

CT has an advantage in evaluating any underlying lung or mediastinal pathology. In empyemas, the pleurae characteristically are smooth and diffusely thickened and enhance after intravenous contrast injection.

Procedure

A large effusion can be drained without image guidance, but smaller or multiloculated effusions are better drained by the radiologist. For this procedure, the seated patient leans forward, “hugging” a pillow against the chest to bring the arms forward and clear the scapulae from the back. A small stool placed under the feet makes the patient feel more comfortable, and the optimum skin position is marked, just above a rib to avoid the intercostal vessels and nerves. A sterile technique with local anesthetic is used. The size of the drainage catheter varies, ranging between small pigtail catheters and large catheters for empyemas.

The depth of the effusion can be judged by both ultrasound imaging and with use of the local anesthetic needle. The catheter can be positioned by means of a Seldinger technique or with a single-step procedure. The latter is often used and involves advancing the catheter with its central stylet into the pleural space. A slight “give” is felt when the pleura is breached, and at this point the catheter is advanced into the effusion while the central stylet is simultaneously withdrawn. The catheter is connected rapidly to a drainage bag and a three-way tap. Samples for cytologic, microbiologic, and biochemical analysis are taken. If drainage is for a short time (24 to 48 hours), use of a bag will normally suffice. If drainage is required for a longer duration (normal with empyemas) or if the effusion is large—presumably because of active fluid production, the drainage bag is replaced by an underwater drainage system.

If aspiration is undertaken, no more than 1.5 L of fluid should be removed at any one time because of the risk of reexpansion pulmonary edema. Larger catheters can be used, but smaller catheters usually are adequate and more comfortable. The catheters should be securely fastened to the skin, and regular saline irrigation should be performed to maintain patency (20 mL of saline every 4 to 6 hours). In some patients, intrapleural fibrinolytics (e.g., streptokinase) given early may improve drainage but are not routinely used, because use of such agents has not been found to reduce long-term mortality or complication rates and may be associated with adverse reactions.

Complications

Complications are few, provided that no bleeding disorder is present. Pneumothorax occurs in less than 5% of cases. Puncture of other viscera can be avoided by inserting the catheter under ultrasound guidance and with blunt dissection through the chest wall to avoid traumatizing the catheter. Pain may occur at the site of insertion, but this is minimized by inserting the catheter laterally for patient comfort. Reactions to streptokinase are much fewer when the purified form is used but previously have included anaphylaxis and bleeding, both intrapleurally and systemically, with larger doses.

Pitfalls and Controversies

The routine use of intrapleural streptokinase has still not been widely adopted because of the expense and the complications associated with the use of the less purified form and the occasional occurrence of large hemorrhages. Its efficacy in the overall management of empyemas remains uncertain.