ultrasound-guided biopsies of the abdomen · pdf fileultrasound-guided biopsies of the abdomen...

Ultrasound-Guided Biopsies of the AbdomenNadia Juliet Khati, MD, Joseph Gorodenker, MD, and Michael C. Hill, MB

Abstract: Ultrasound is the most useful imaging technique for guid-ing biopsies and fine-needle aspirations that are performed percu-taneously. It is a safe and accurate method to guide needles usingreal-time imaging into abdominal and pelvic organs and masses. Itsadvantages over computed tomographyYguided biopsies are that itcan be done portably, is less expensive, and does not use ionizingradiation. Even lesions as small as 1 cm can be biopsied assuming theprocedure can be performed safely and the mass can be visualizedsonographically.

This review article discusses the different components of per-forming percutaneous biopsies in adult patients including review ofthe patients imaging studies, coagulation status, and medications. Weexplain when and how we do core biopsies versus fine-needle aspi-rations to obtain a diagnosis depending on the suspected diagnosisand organ/mass being biopsied. We also describe how we take careof the patient after biopsy.

Ultrasound can be used safely and successfully to perform biop-sies and fine-needle aspirations in solid organs of the abdomen andpelvis including the retroperitoneum and abdominal wall. This tech-nique allows adequate tissue sampling with real-time monitoring du-ring placement of the biopsy needle. Postprocedure complications areextremely rare with the vast majority of our patients requiring only ashort recovery time of 2 hours following the biopsy.

Key Words: ultrasound, biopsy, fine-needle aspiration,ultrasound-guided biopsies, biopsy abdomen, biopsy pelvis

(Ultrasound Quarterly 2011;27:255Y268)

U ltrasound is the most useful imaging technique for guid-ing percutaneous and endoscopic biopsies. It is safe and

accurate and allows one to guide a needle in a real-time man-ner into organs, masses, and lymph nodes.1,2 NolsLe et al,2 in areview of almost 8000 ultrasound-guided percutaneous pro-cedures, found a complication rate of only 0.187% and a mor-tality rate of 0.038%. Even small masses down to 1.0 cm canbe successfully biopsied depending on their location and so-nographic visibility.1,3 Its major advantage over computed

tomography (CT) is that it is portable, allows real-time visu-alization of the biopsy needle and its tip, is less expensive, andmost importantly does not use ionizing radiation.4 Its greatestdisadvantage is that that it is more operator dependent. Also,some lesions or organs are difficult to see and may not beamenable to biopsy because of the patient’s body habitus orintervening structures such as air or bone. In those cases, CTguidance might be preferable.

The component parts in performing a biopsy are asfollows:1. Review the clinical indications for performing the biopsy

to make sure it is indicated. Look at prior imaging studiesif available to ascertain if the biopsy can be performedsafely using ultrasound guidance.

2. Ensure the coagulation status of the patient is within arange that allows the biopsy to be performed safely. If not,consult with the patient’s physician, including a hema-tologist if necessary, to see if the abnormal values can becorrected.

3. Properly obtain informed consent.4. Perform the biopsy following the appropriate administra-

tion of local anesthesia.5. Perform care of the patient after biopsy.6. Review quality assurance of pathology reports to deter-

mine the percentage of failed biopsies.

Coagulation Status of the PatientIn this section of the article, we describe our approach

to screening patients by obtaining coagulation status beforeall of our procedures. We acknowledge that this is a complexand controversial issue with the literature providing ratherconfusing guidelines. This is why different medical institu-tions may have their own standards for performing screeningcoagulation tests on their patients based on their experiencewith complications.

When a procedure is scheduled in our ultrasound de-partment, we first review the clinical indications and imagingstudies to ensure that the procedure is both clinically indicatedand safe to perform. We then have to assess the coagulationstatus of the patient to minimize the most feared complicationrelated to the procedure, namely, postbiopsy hemorrhage. Be-cause a large number of our biopsies are scheduled on anoutpatient basis, we as radiologists have very little clinical in-formation on these patients other than the history relevant tothe procedure itself. For patients who are scheduled for a solidorgan biopsy (liver, kidney, spleen, adrenal, and pancreas), weroutinely screen them by requesting a complete blood count,a prothrombin time (PT) and international normalized ratio(INR), and activated partial thromboplastin (aPTT). We do notrequire repeat coagulation tests if results are available fromthe past 3 months unless the patient’s clinical status demands

REVIEWARTICLE

Ultrasound Quarterly & Volume 27, Number 4, December 2011 www.ultrasound-quarterly.com 255

Received for publication April 8, 2011; accepted August 29, 2011.Department of Radiology, The George Washington University Medical Center,

Washington, DC.The authors declare no conflict of interest.Reprints: Nadia Juliet Khati, MD, Department of Radiology, The George

Washington University Medical Center, 900 23rd St, NW Washington, DC20037 (e-mail: [email protected]).

Supplemental digital contents are available for this article. Direct URL citationsappear in the printed text and are provided in the HTML and PDF versionsof this article on the journal’s Web site (www.ultrasound-quarterly.com).

Copyright * 2011 by Lippincott Williams & Wilkins

Copyright © 2011 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

it (history of lupus, chronic use of anticoagulation therapy,patients with known bleeding disorders). As mentioned at thebeginning of this section, we recognize that ordering coagu-lation tests is at the discretion of the physician performing theprocedure. Malloy et al5 recently published extensive detailedguidelines for coagulation status and management based onthe risk of bleeding of a large group of vascular and nonvas-cular percutaneous procedures. In general, they classify intra-abdominal and transabdominal biopsies as ‘‘moderate risk’’for bleeding, whereas renal biopsies are considered to havea more ‘‘significant risk’’ of bleeding.5 Their consensus panelconcluded that only INR and aPTT (for patients on heparin)would be necessary for the moderate risk for bleeding group.They recommended adding a platelet count and hematocritlevel for the significant risk group. We also obtain a list of thepatients’ medications, paying specific attentions to ones thatmay alter coagulation. This allows correct interpretation ofall test results and a better assessment of comorbidities alongwith avoiding unnecessary rescheduling or cancellation ofprocedures.

Our accepted cutoff numbers for each of the coagulationparameters are summarized in Table 1. It should be empha-sized that these are general guidelines that we follow; however,they can be altered on a case-by-case basis depending on thepatient’s clinical condition. Based on a wide review of the lit-erature and their experience, O’Connor et al6 have suggestedcutoff INR values of up to 3.0 for procedures performed usinga 20-gauge needle and up to 2.0 for most other proceduresincluding tissue biopsies, transhepatic cholangiograms, andplacement of percutaneous catheters. They consider a safeplatelet count level of 25,000/KL or greater for soft tissuebiopsies and other procedures and have no specific cutoffvalue when doing FNAs using a 20-gauge needle. It should beremembered that a platelet count is only a number and doesnot give us information regarding platelet function. Some ad-vocate the use of the bleeding time to assess platelet function;however, it has limited sensitivity and specificity in predict-ing the risks of hemorrhage related to invasive procedures. Forthis reason, we do not routinely use the bleeding time whenperforming our procedures. Others consider safe criteria de-fined as a platelet count of 50,000/KL and an INR of less than1.6.7 Caturelli et al8 performed 229 liver punctures under ul-trasound guidance on 85 patients using fine needles (20- and22-gauge) for diagnostic purposes or therapeutic injectionof ethanol. Their patients had platelet levels of less than50,000/mm3 and/or PT times that were less than 50% of con-trols, and none had any bleeding complications. In addition,none of their patients received any type of blood products tocorrect their coagulation parameters.8

Interpretation of the PT, INR, and aPTT results is per-formed in conjunction with any relevant history pertaining tochronic liver disease, malabsorption syndromes, and malnu-trition. A large number of our scheduled liver biopsy patientshave underlying liver disease, most commonly hepatitis andcirrhosis, and any of these conditions may abnormally prolongcoagulation parameters, particularly the PT. Knowledge of thepatients’ prescribed medication specifically the intake of as-pirin, clopidogrel (Plavix; Bristol-Myers Squibb, New York,NY), warfarin, or heparin is crucial to the correct interpreta-tion of homeostasis. Aspirin and clopidogrel are both plate-let inhibitors, and we instruct patients to stop taking them 5 to7 days before the procedure. This is another topic that is upfor debate according to the literature and varies from oneinstitution to the next. Whereas Winter et al9 seem to be inagreement with our recommendations, Malloy et al5 haveslightly different guidelines. They advise withholding clopi-dogrel 5 days before procedures that are both in the moderaterisk and significant risk for bleeding, whereas aspirin isstopped only 5 days before a procedure in the last groups ofpatients. On the other hand, Atwell et al7 studied the overallincidence of significant hemorrhage following 15,181 percu-taneous image-guided biopsies from a variety of solid organsin patients taking aspirin within 10 days of the procedure.They concluded that overall the risk of major hemorrhage wasnot significantly increased in patients on aspirin (0.6%) com-pared with those not on aspirin (0.4%).

Withholding clopidogrel is relatively easy for the groupof patients who take these drugs as a preventive measure, thatis, to avoid any cardiovascular events. However, patients whohave coronary artery stents or a history of prior heart attackor stroke and those with peripheral artery disease reportedlyhave a 3 times higher risk of developing stent thrombosis, tran-sient ischemic attacks, and ischemic strokes if they discontinuetheir aspirin intake.10,11 When it can be stopped safely, clo-pidogrel should be restarted as soon as possible after the pro-cedure using a loading dose.12 All of this should be done afterconsulting with the referring physician to ensure that the ben-efits of the procedure outweigh the risks of discontinuingclopidogrel, even temporarily.

Patients taking either warfarin or heparin are the onesmost likely to have bleeding complications following a proce-dure. Warfarin inhibits the formation of vitamin KYdependentcoagulation factors, thus prolonging the INR and PT, whereasheparin inhibits coagulation factors of the intrinsic pathwayprolonging the aPTT. Low-molecular-weight heparin (LMWH)has antiYXa factor activity and, to a lesser degree, inhibitionof all the coagulation factors of the intrinsic pathway; how-ever, it does not alter the aPTT. We instruct our patients to stopwarfarin 3 to 5 days before a procedure and advise bridginganticoagulation therapy with heparin or LMWH for those withthe greatest risks of thromboembolic accidents. Alternatively,warfarin dosage should be decreased to adjust the INR leveldown to 1.5 to 2. When using heparin or LMWH bridgingtherapy, they should be stopped 6 and 12 hours, respectively,before the biopsy.

For nonsteroidal anti-inflammatory drugs (NSAIDs)other than aspirin, patients are instructed to take them as di-rected by their treating physicians. Although some NSAIDs

TABLE 1. Normal and Cutoff Values for Coagulation Factorsfor Patients Undergoing Organ BiopsiesTest Reference Ranges Cutoff in Our Practice

PT 10Y13 s 10Y13 s

INR 0.9Y1.1 G1.5

aPTT 25Y35 s 25Y35 s

Platelets 150,000Y450,000/KL 950,000/KL

Khati et al Ultrasound Quarterly & Volume 27, Number 4, December 2011

256 www.ultrasound-quarterly.com * 2011 Lippincott Williams & Wilkins


have antiplatelet aggregation function, they are short-actingdrugs, and the effect is quickly reversible compared with mostother anticoagulants. Table 2 shows the guidelines on how wemanage patients on anticoagulation therapy in our departmentbefore scheduling a biopsy.

When the hemostatic status is compromised and the PT,aPTT, and INR cannot be corrected by adjusting anticoag-ulation therapy, patients should be given the option of havingeither platelet transfusions or fresh frozen plasma (FFP). Weusually recommend using 2 to 4 units of platelets or FFP ap-proximately 1 hour before the procedure, with preferably thelast unit being transfused during the biopsy. We do not rou-tinely recheck the INR, PT, and aPTT after these transfusions.According to the literature, however, there are no set guide-lines to determine the exact dose of FFP needed to correct onlymildly elevated INR values, and most of the time, the numberof units used is insufficient to correct it.13 Abdel-Wahab et al14

evaluated the effect of FFP on 121 patients using pretransfu-sion and posttransfusion PT and INR values 8 hours after trans-fusion. Patients had pretransfusion PTs that varied between13.1 and 17 seconds and INR values between 1.1 and 1.85.They concluded that FFP only partially corrected PT valueswhen they were mildly elevated with failure to normalize thePT in 99% of cases.14 In addition, FFP and platelet transfu-sions carry a risk of severe reactions including transfusion-related acute lung injury, volume overload, infection, andanaphylactic reactions.6When confronted with a situation wherethe patient would require blood product transfusions to correcttheir coagulopathic state, we generally discuss the different op-tions with the referring physician, making sure that we care-fully weigh the risks of the procedure against the benefits ofobtaining a pathological diagnosis.

PERFORMANCE OF THE BIOPSY PROCEDUREBefore each procedure, patients are consented by the ra-

diologist performing the biopsy. We explain every step of theprocess and describe the potential complications associatedwith each individual biopsy and the alternatives to the biopsy.

We make sure that the patient has either somebody availableto drive him/her home. Patients are required to sign a consentform, which is witnessed by the sonographer or nurse avail-able at the time. In addition to this, a ‘‘time out’’ is called bythe sonographer with the radiologists in attendance beforethe procedure itself, at which time the patient’s identity andtype of biopsy are verified.

We no longer have patients fast and in fact encouragethem to have a light meal before the procedure. Before startingany biopsy, we make sure the patient fully understands the pro-cedure. Reassuring the patient and keeping them informedthroughout the process are very important. This helps relieveanxiety and apprehension, and using these techniques, we havenoted anecdotally a reduction in the number of vasovagalreactions that occur.

Before starting the biopsy, all the necessary supplies toperform the procedure efficiently in the least amount of timeshould be readily available in the room. We use a 5- to 7-mHzcurved array probe for deep biopsies to which a biopsy guidecan be attached. For more superficial biopsies, we use a 10- to12-mHz linear array probe and a freehand technique. Thetransducer face of the probe is covered with some ultrasoundgel and placed into a sterile plastic sheath. A rubber band isput around the transducer face to keep the gel in place, and asmall amount of sterile gel is used to provide good acousticcontact between the sheathed transducer and the patient’s skin.Some authors prefer not to use the plastic sheath as it maydegrade the image.1 They rely on cleaning the transducer withpovidone-iodine (Betadine), which works well at preventinginfection.15

Over the years, we have gone away from using con-scious sedation for performing biopsies and rely on the liberaluse of local anesthesia (1% lidocaine). We tell patients thatthe biopsy procedures that we do are similar to going to thedentist. Although the administration of the lidocaine initiallydoes hurt, once it is fully administered, patients should feelno pain. The interventional radiologists in our department useconscious sedation as their procedures are more complex andlengthy, whereas our biopsy takes only minutes to perform oncethe local anesthesia has been administered. Mueller et al16

showed that patients having visceral biopsies experiencedmore pain with local anesthesia alone versus local anesthesiaand conscious sedation. The satisfaction scores, however, weresimilar for each group. The administration of conscious seda-tion lengthens the procedure and requires constant monitor-ing of the patient. Patients also may act differently and may notobey simple commands such as ‘‘hold your breath,’’ which maybe required for some liver and kidney biopsies.

The skin entry site should be cleansed and draped in asterile fashion and should not be close to a scar as this canmake it difficult to advance the biopsy needle. Our biopsytrays contain either povidone-iodine (Betadine) or chlorhexi-dine as the topical antiseptic agent. We have no specific pre-ference for using one over the other. The surgical literatureseems to be divided as to their efficacy in preventing infec-tion, with some stating that chlorhexidine is comparable topovidone-iodine (Betadine).17 Others conclude that the formerperforms better than the latter as a preoperative antiseptic.18,19

Advantages of chlorhexidine include that it does not cause

TABLE 2. Guidelines Used in Our Department for ManagingPatients on Medications That May Affect CoagulationMedication/Anticoagulants Our Guidelines

NSAIDs Do not need to stop

Aspirin Stop 5Y7 d before procedure

Clopidogrel Stop 5Y7 d before procedure when safe for the patient(no stents)*. Restart immediately after with aloading dose

Warfarin Stop 3Y5 d before procedure or taper dose to get INRat 1.5 on day of procedure and restart original doseafter procedure, or stop 3Y5 d before procedure withheparin or LMWH bridge before and after procedurefor 2 d. Restart warfarin after procedure

Heparin Stop 6 h before procedure and restart 6Y12 h after

LMWH Stop 12 h before procedure and restart 12 h after

*Patients with stents cannot stop clopidogrel safely if they are within 2 months ofbare-metal stent (nonYdrug coated) placement or within 6 to 12 months of drug-elutingstent placement.

Ultrasound Quarterly & Volume 27, Number 4, December 2011 Ultrasound-Guided Biopsies of the Abdomen

* 2011 Lippincott Williams & Wilkins www.ultrasound-quarterly.com 257


itching and does not leave a brownish stain on the skin likeBetadine.19

In our patients who are all adults, we use up to 25 mL of1% lidocaine (10 mg/mL) to achieve complete local anesthe-sia. The maximum dosage of lidocaine 1% that can be admi-nistered is 4.5 mg/kg not to exceed 300 mg. Local anesthesiais achieved for up to 30 minutes or more using this dosage.Adverse effects are very rare and involve the central nervoussystem (disorientation, convulsions, respiratory depression,coma), cardiovascular system (bradycardia, cardiovascular col-lapse), and even more rarely allergic reactions. It is importantto make sure the skin is properly anesthetized by injecting 2 to3 mL of lidocaine intradermally (not subdermally) by makinga wheal. This is done using a 25-gauge, quarter-inch needle.The soft tissues down to the biopsy site are then anesthetizedusing ultrasound guidance and a 20- or 22-gauge spinal nee-dle. As the needle is advanced, the plunger of the syringeshould be drawn back before injecting to make sure the needletip is not in a blood vessel. Patients usually do not experiencemuch pain as the needle goes through fatty tissue. When theneedle traverses the muscular layer, however, the patients mayexperience some pain. They should be forewarned of this asone sees the needle-tip approach the muscle layer on the ul-trasound monitor (see Video, Supplemental Digital Content[SDC] 1, http://links.lww.com/RUQ/A48 which demonstratesdeep injection of lidocaine to the level of the biopsy site). Ifthe chosen biopsy tract traverses into the peritoneum, wemake sure to administer 3 to 5 mL of lidocaine directly onthe peritoneal surface before advancing the needle throughit. The pain elicited by advancing through the peritoneumthat is not properly anesthetized can lead to a vasovagalreaction. Once we have achieved good local anesthesia, weremove the needle and make a small 3-mm nick in the skinto allow easy movement of the needle. This nick is madebystretching the skin transversely and introducing a scal-pel blade at a 45-degree angle into the skin, with the sharpedge facing upward. The skin is then cut in an upward fash-ion so as to avoid cutting into the deeper soft tissues. Thebiopsy needle is gently pushed through the nick and usingultrasound guidance is advanced through the anesthetizedsoft tissues down to the organ or mass on which to performa biopsy.

If there is difficulty visualizing the needle tip, we movethe focal zone of the transducer to its expected position. Wealso adjust the acoustic output appropriately, as seeing the nee-dle tip, particularly in echogenic soft tissues can be especiallydifficult.20 Moving the needle tip to and fro (‘‘pump man-euver’’) is also helpful as is the use of color flow Doppler.20Y22

This technique is especially useful when using fine needlesand a steep angle.22 If the needle has a stylet, moving this backand forth in the barrel of the needle improves needle-tip vis-ibility.22 If the target organ moves with respiration, breathingshould be suspended for 3 to 5 seconds while performing thebiopsy. It is often useful to have a few practice runs doing thisto make sure the patient can correctly follow the command. Weusually take a video clip documenting the biopsy needle en-tering and exiting the tissue being biopsied (see Video, SDC 2,http://links.lww.com/RUQ/A49, which demonstrates the biopsyneedle entering and exiting the liver).

In cases requiring a coaxial technique, we use the co-axial needle to administer the lidocaine to the level of biopsysite using ultrasound guidance. The tip of the needle is usuallyeasily seen while the lidocaine is administered as it producesa swirl of echoes in the soft tissues around the tip. The lido-caine left in the needle is removed by fully inserting the blunt-tipped trocar before inserting the biopsy needle.

Most of the abdominal and pelvic biopsies are corebiopsies, done using an automated, spring-loaded, 18-gaugecutting needle that has a variable throw (13, 23, or 33 mm).Overall, they provide a diagnosis up to 94% of the time.23Y25

Arabas et al26 have shown that a 14- to 18-gauge cutting nee-dle is better in providing a diagnosis than a 20-gauge cuttingneedle. In 841 patients, the larger needle provided a patho-logical diagnosis in 96.8% versus 84.1% for the smaller nee-dle. The needle can be introduced directly to the target site byitself or through a 17-gauge coaxial needle, which we favor.The core specimens obtained are placed in 10% formalin so-lution or on a Telfa pad moistened with nonbacteriostatic sa-line. If a smear is needed for cytopathology, the core can beplaced on a glass slide, and a second glass slide can be gentlybrushed over the specimen (touch prep). This will depositsome cellular materials on the glass slide, which can then besmeared and stained for cytopathologic analysis.

In cases where lymphoma is the possible diagnosis, wedo FNAs. At our institution, FNAs are widely used for thediagnosis of hematologic neoplasms. Our cytopathologists per-form flow cytometry to detect cell markers (immunopheno-typing) on a small sample of cells. This allows accuratedifferential diagnosis between reactive processes and subclas-sification of B-cell non-Hodgkin lymphomas and T-cell non-Hodgkin lymphomas. We use 22-gauge needles for the FNAsin addition to core biopsies, using a coaxial technique. TheFNA needles should be long enough to extend out beyond thetip of the coaxial needle.27 The 17-gauge coaxial needle is mucheasier to visualize sonographically than the 22-gauge needle andis less likely to bend out of the field of view. The tip of the fine-needle that is beyond the tip of the coaxial needle is easily seenwithin the mass being sampled because of its back-and-forthmotion (Fig. 1, seeVideo,SDC3,http://links.lww.com/RUQ/A50,which demonstrates FNAs of the small focal liver lesion seen onthe CT scan. Lymphoma was the suspected diagnosis). Beforestarting the sampling process, we attach an air-filled syringeto the needle, which acts as a handle for the back-and-forthmotion needed to perform the FNA The air-filled syringe isthen used to expel the cellular contents in the barrel of theneedle without having to separately attach/disattach the nee-dle. Our cytopathologists prefer us not to apply suction to thesyringe, and so we use a nonaspiration technique that providesdiagnostically superior specimens.28,29 This technique worksbetter with larger-bore needles.27 The cellular material isdrawn into the barrel of the needle by capillary action.29 Theuse of suction tends to draw blood into the needle, makingit difficult to visualize the native cells. We do a minimum of3 FNAs (to ensure adequacy of sampling because our cyto-pathologists are not always present during the procedure),which are smeared onto slides for cytopathology and thenrinsed out of the needle into RPMI medium for flow cyto-metry. After the FNAs, we also perform at least 3 core biopsies




through the coaxial needle. These additional cores are per-formed as a backup for immunohistochemistry or in caseswhere diagnoses other than lymphoma are in the differential.

LIVER BIOPSYThe usual indications for performing a liver biopsy are

for diagnosing hepatocellular disease and the etiology of ahepatic mass30 (Table 3).

A biopsy should be performed on coagulopathic patientswho need a biopsy because of diffuse hepatocellular disease,using the transjugular approach.31,32 When this is not a pro-blem, we do our liver biopsies using a subcostal approachin the midline epigastrium, taking our core from the left lobeof the liver (see Video, SDC 4, http://links.lww.com/RUQ/A51,showing a biopsy of the left lobe of the liver using a short-throw technique in a patient with sarcoidosis). If the left lobeis small and not easily amenable, we place the patient in the

left lateral decubitus position and use a subcostal approach.Less preferably, we would use the intercostal route. We makesure to administer 3 to 5 mL of lidocaine directly onto the liversurface using ultrasound guidance. To limit the number ofpasses to one, we use an automated, spring-loaded, 18-gaugecutting needle with a throw of 33 mm that provides a nice longspecimen (see Video, SDC 5, http://links.lww.com/RUQ/A52,which demonstrates the technique for a core biopsy from theleft lobe of the liver using a deep-throw technique).23,27

For biopsy of a liver mass, we usually use a coaxial tech-nique3,26 (see Videos, SDC 6, http://links.lww.com/RUQ/A53,which demonstrates FNAs of a small focal liver lesion, andSDC 7, http://links.lww.com/RUQ/A54, which demonstratessubsequent core biopsy of the same liver lesion). Advantagesof this technique over the noncoaxial method for focal lesionsare that multiple samples can be obtained once the coaxialneedle is in place and the capsule of the organ is pierced onlyonce. There is also evidence in the literature that this methoddecreases the risk of tumor seeding along the needle track, andthis is crucial in potential liver transplant recipients.33 Maturenet al33 found a 0% rate of tumor seeding in 101 patients withhepatocellular carcinoma using a 17-gauge coaxial system andan 18-gauge biopsy cutting needle. Hatfield et al34 comparedthe complication rate of performing liver and renal biopsiesusing imaging guidance with a coaxial system (764 biopsies)and without a coaxial needle (296 biopsies). They found nodifferences between the 2 methods, with minor complicationsoccurring in 3.4% and 2.6% of the cases when using the non-coaxial and the coaxial techniques, respectively. Major com-plications requiring blood transfusions or surgery occurred in1.0% of noncoaxial and 0.9% of coaxial techniques.

If the liver is cirrhotic, we examine the expected path ofthe needle using duplex and color flow Doppler to avoid anyportosystemic venous collaterals that might be in the way. Wetry to perform a biopsy on all masses by advancing the coaxialneedle through a cuff of normal liver tissue down to the surfaceof the mass when possible. Taking into account the size of thelesion, we take our sample from the outer one third of the mass

TABLE 3. Indications for Liver Biopsy

Diffuse hepatocellular disease

Alcoholic liver disease

Nonalcoholic hepatic steatosis

Viral hepatitis

Autoimmune hepatitis

Heavy metal storage disordersVhemochromatosis, Wilson disease

Unexplained liver function test elevation

Unexplained cholestasis

Use of hepatotoxic drugsVmethotrexate

Unexplained Systemic IllnessVFUO, granulomatous disease

Unexplained hepatosplenomegaly

Suspected liver transplant rejection

Liver transplant donor

Focal liver disease

PrimaryVhepatocellular carcinoma, cholangiocarcinoma

SecondaryVmetastatic disease

FUO indicates fever of unkwown origin.

FIGURE 1. Contrast-enhanced CT scan of the abdomenshowing a small well-defined hypodense solid lesion in themedial segment of the left lobe (arrow) in a patient withsuspected lymphoma.

FIGURE 2. Sagittal transabdominal ultrasound showing thebiopsy procedure performed through the ascites with theneedle tip into the left lobe of the liver (arrow).




and avoid the center, which may have pathologically non-diagnostic necrotic tissue. If the mass is 2.0 cm or less, we takeour specimen from its center. The ability to perform biopsieson small masses depends not only on their conspicuity andskill of the operator but also on the patient’s ability to holdtheir breath.

The presence of ascites is not a contraindication tobiopsy, although it can make it more difficult if it is markedin amount (Fig. 2). This is especially true when trying to sam-ple a mass deep in a cirrhotic liver (Fig. 3, see Video, SDC 8,

http://links.lww.com/RUQ/A55, which demonstrates a core bi-opsy performed on a deep liver lesion in a patient with a cir-rhotic liver and ascites). In some cases, paracentesis before theprocedure may be helpful.

Liver biopsies are very safe to perform when the patient’sblood clotting mechanism is normal. In a series of 715 patientshaving liver biopsies, there were only 2 major complicationsbecause of hemorrhage (Fig. 4) and no deaths.30 The mostfrequent minor complication was pain. This usually occurs atthe time of the biopsy or shortly thereafter. The pain is usuallymild but can be severe and is presumably due to blood or bilestretching the liver capsule35 or intraperitoneal bleeding irri-tating not only the peritoneum but also the intestinal tract. Notinfrequently, the pain can be referred to the shoulder, often theleft. This pain is usually self-limiting and goes away within2 hours. In some cases, we prescribe acetaminophen by mouth,but only rarely do we have to resort to intramuscular or in-travascular injection of pain medication.

A pathological diagnosis can be made in 99.4% ofbiopsies performed for diffuse hepatocellular disease.30 Theaccuracy rate for diagnosing even small liver masses has in-creased from 79% in 1987 to 99% in 1999.1,3

PANCREATIC BIOPSYNowadays, most pancreatic mass biopsies are performed

using endoscopic ultrasound, which has largely replaced thepercutaneous approach as both techniques produce similar re-sults.36 In cases where the endoscopic approach is not tech-nically feasible or where it fails to yield a diagnosis, thepercutaneous approach can be used. This is usually done toconfirm the diagnosis of an unresectable pancreatic cancer.Masses in the head, body, and proximal tail are biopsied usingan anterior approach through the abdominal wall, with the pa-tient supine (Fig. 5). Filling the stomach with water may al-low better visualization of the mass, and the biopsy can beperformed through-and-through the gastric wall if necessary.37

A loop of small bowel may also be traversed; however, weavoid going through the colon, although others have used thisapproach.38 A mass in the tail may be better approached

FIGURE 4. A, Sagittal abdominal ultrasound of a liver mass showing its subcapsular location (arrow). The biopsy track is shown.B, Contrast-enhanced CT scan of the abdomen in the same patient. Following the biopsy, the patient developed intractablebleeding around the lesion (arrow). C, Contrast-enhanced CT scan of the abdomen in the same patient a few days later. He requiredarterial embolization with subsequent infarction of the spleen.

FIGURE 3. Transabdominal ultrasound of the liver in apatient with ascites and a slightly nodular liver. There is awell-defined heterogeneous solid right liver lobe lesion (arrows).




through a left intercostal space, with the patient in the rightdecubitus position. Care must be taken to avoid the adjacentspleen and left colon.

We use a core needle and a coaxial technique. Beforeadvancing the needle, color flow Doppler imaging should beperformed to avoid any intervening vessels.39 This is espe-cially advised in cases where the portal vein or splenic veinis occluded as there may be many small venous collateralsaround the pancreas.

Percutaneous pancreatic biopsies are probably morerisky than liver or kidney biopsies as the pancreas has manynormal surrounding blood vessels, which need to be avoided.That is why biopsy of a liver metastasis, if one is present, isfavored over biopsy of the primary pancreatic mass.40 Biop-sies of liver metastases also allow staging of the pancreaticdisease. The other problem with making a percutaneous di-agnosis is the presence of acute pancreatitis in associationwith the cancer. This may lead to sampling the area of in-flammation rather than the tumoral tissue. Chronic pancreati-tis can also be a problem as its symptoms can mimic that ofpancreatic cancer. In a study of 81 patients, 54 with pancreaticcancer and 27 with chronic pancreatitis, DelMaschio et al41

demonstrated that FNA biopsies were able to differentiatebetween the 2 with an accuracy of 94%.

The accuracy rate of ultrasound-guided percutaneousbiopsy reported by Paulsen et al39 was around 92.5%, and thesensitivity was 93.3%. Other authors have produced similarresults.25,38,42Y45 Some results using CT guidance show alesser success rate; however, this could be due to the moreaccessible masses that are easier to have a biopsy performedon being triaged to ultrasound.38,46 The size of the mass alsoaffects obtaining a diagnosis. Brandt et al38 had a success rateof 92% for masses larger than 3 cm, whereas the rate fell to81% for lesions smaller than 3 cm. In the same study, theydemonstrated a better success rate using a larger needle (92%for 16- to 19-gauge cutting needles vs 85% for 20- to 22-gauge

needles). They had better success for masses in the body/tail(93%) compared with the head (84%). The complication rateof percutaneous pancreatic biopsies (1.1%Y2.6%) can be ma-jor or minor.38,39,44 Minor complications include vasovagalreactions, pain requiring narcotics, and small bleeds. Majorcomplications include bleeding that requires treatment andacute pancreatitis (2.6%). There does not appear to be anysignificant difference in the rate of complications betweenFNAs and core biopsies.39 Seeding of the needle tract is a rarebut possible complication of percutaneous fine-needle biopsiesof primary pancreatic neoplasms or pancreatic liver metas-tases.47,48 Kosugi et al49 performed biopsies of 73 pancreaticneoplasms using 21-gauge needles. Each lesion required 1 or2 samples, and the needles were rinsed in 10% alcohol whena second puncture was performed. Tumor implantation alongthe needle tract was found in only one case 2 months after theprocedure in a patient with a moderately differentiated carci-noma. This was subsequently successfully removed with noadditional recurrence.49

SPLENIC BIOPSYThis is usually performed to ascertain the pathology of

a splenic mass, usually lymphoma or a metastatic deposit(Table 4). The procedure is carried out using a left intercostalapproach, with the patient in the right lateral decubitus posi-tion (Fig. 6). If the spleen/mass is large, a subcostal approachmay be possible. If lymphoma is the suspected diagnosis, weperform FNAs and core biopsies using a coaxial technique asdescribed earlier in this article. We do not perform FNAs iflymphoma is not in the differential diagnosis. Keogan et al50

recommend traversing as little splenic tissue as possible to getto the mass, which is the opposite of what is recommended forbiopsy of liver masses.

Splenic biopsies are as safe to perform as liver biopsies,assuming the patient is not coagulopathic.50,51 In fact, thediagnostic accuracy and major complication rates are com-parable to those reported for renal and liver biopsies whenusing 18-gauge or smaller needles.52 The most common majorcomplication when it occurs is hemorrhage (Fig. 7), whichmay require embolization or even emergency splenectomy.53

TABLE 4. Indications for Splenic Biopsy

Benign masses

Hemangioma

Lymphangioma

Hamartoma

AbscessesVpyogenic, fungal, Pneumocystis carinii, MAI, TB

Inflammatory processes

Sarcoidosis

Inflammatory pseudotumor

Peliosis

Malignant masses

Primary neoplasmsVlymphoma, angiosarcoma

Secondary neoplasmsVmetastases from melanoma, breast, lung, ovarian,colon, gastrointestinal

MAI indicates mycobacterium avium-intracellulare; TB, tuberculosis.

FIGURE 5. Transabdominal ultrasound of the pancreasshowing a large solid hypoechoic mass in the tail (M).




The accuracy of splenic biopsy in making a pathological di-agnosis is around 84% to 91%.50,53,54

RENAL BIOPSYRenal biopsies are performed to determine the etiology

of renal parenchymal disease and the pathology of renalmasses (Table 5).55,56 The procedure is performed with thepatient in a prone position. A pillow is placed under the ab-domen in thin patients. The biopsy is taken from the lowerpole of either kidney to obtain cortical tissue that contains glo-meruli. An initial scan is done to evaluate the kidneys’size andechogenicity and to determine which lower pole is most ame-nable to biopsy. In some obese patients, the prone position willnot work as the kidney may be poorly visualized or be too deepto have a biopsy performed on. In that case, the decubitusposition has to be used. In this position, the kidney moves bothinferiorly and anteriorly, placing it closer to the skin surface.

If the kidney is still not adequately visualized, an intercostalapproach in the prone position can be attempted. This ap-proach is technically more difficult than the other two. Visu-alization of the kidney is limited, making it more likely thecore needle will enter the renal sinus and damage the renalcollecting system or vasculature.

We do renal biopsies using an automated, spring-loaded,18-gauge cutting needle and a coaxial technique.23,27 Thetip of the coaxial needle is guided down to the outer cap-sule of the lower pole of the kidney (see Video, SDC 9,http://links.lww.com/RUQ/A56, which demonstrates core bi-opsy of the lower pole of the kidney in a patient with im-munoglobulin A nephropathy). The biopsy is performed insuspended respiration, and 3 to 4 specimens are obtained mea-suring 1.3 cm in length. These are placed on individual Telfapads moistened with nonbacteriostatic normal saline and givendirectly to the pathologist for processing. The need for 3 to 4cores is dictated by the required number of glomeruli perspecimen to make an accurate diagnosis. The nephrologistsat our institution prefer to have at least 15 to 20 glomeruliper core. Of the 4 cores obtained, 2 are used for histology:

FIGURE 6. A, Contrast-enhanced CT scan of the abdomen in a patient with history of melanoma with multiple, solid, partiallyenhancing splenic lesions (arrows). B, Transbdominal ultrasound of the spleen showing the biopsy tract for one of the spleniclesions. The diagnosis was metastatic melanoma.

FIGURE 7. Contrast-enhanced CT scan of the abdomenfollowing splenic mass biopsy (same patient as Fig. 6). Thepatient developed a small postprocedure hemorrhage(thin arrow). The curved arrow denotes a large left adrenalgland metastatic lesion.

TABLE 5. Indications for Renal Biopsy

Parenchymal renal disease

Unexplained renal failure

Nephrotic syndrome

Unexplained hematuria

Systemic disease with renal dysfunction : systemic lupus erythematosis

Goodpasture syndrome

Wegener granulomatosis

Renal transplant rejection

Renal mass

Small renal mass amenable to local resection or ablation

Renal mass in renal insufficiency

Renal mass in a solitary kidney

Multiple renal masses

Renal mass in a patient with a known primary malignancy

Renal mass in a patient with metastatic disease




1 for electron microscopy and 1 for immunofluorescencemicroscopy.

The appropriate approach to biopsy depends on the lo-cation of the mass within the kidney and can be determinedonly by scanning the patient. We use the same coaxial techniquethat we use for parenchymal renal disease and usually obtain3 tissue specimens that are placed in formalin (see Video, SDC10, http://links.lww.com/RUQ/A60, which demonstrates a corebiopsy performed on a large left lower pole renal mass in apatient with lymphoma).

Renal biopsies are a safe procedure; however, compli-cations do occur and include hematoma at the biopsy site,hemorrhage that may require embolization, and pseudoan-eurysm formation (Fig. 8).57,58 Maya and Allon59 performedultrasound-guided renal biopsies in 100 outpatients to evaluatethe safety of the procedure. They used 16-gauge needles andobserved their patients for 8 hours following the biopsies with1 or 2 needles passes done in 91 patients and 3 to 4 passes inthe remaining 9 patients. Only 13 developed a small postpro-cedure perinephric hematoma, and 4 were hospitalized over-night because of a drop of greater than 4% in their hematocrit.However, none developed major complications that wouldrequire a blood transfusion or other intervention.59

They produce a pathological diagnosis in up to 100%of patients with parenchymal renal disease and in 82% to100% of renal masses.56Y58,60

ADRENAL BIOPSYSmall adrenal masses are frequently found during CT

and magnetic resonance imaging studies of the abdomen per-formed for other reasons. These are usually nonfunctioning ad-enomas unless there is a history of malignancy.61 Even withthe history of malignancy, the likelihood of metastatic diseaseis only 50% unless the patient has small cell carcinoma ofthe lung. Despite a workup using an adrenal CT or magnetic

resonance imaging protocol or when indicated a CTYpositronemission tomography scan, some of the masses remainindeterminate.62Y65

Biopsy of an adrenal gland mass is usually performedto determine if it is a metastatic lesion in a patient with aknown primary. It is also indicated where there is metastaticdisease, and the primary could be adrenocortical carcinoma.If a pheochromocytoma is suspected based on the clinicalsymptoms or imaging findings, biochemical testing (ie, mea-surement of fractionated metanephrines and catecholaminesin a 24-hour urine specimen) should be done before perform-ing a biopsy because of the higher risk of bleeding and hy-pertensive crises.66Y68 Most of these biopsies are performedusing CT guidance as many of these lesions cannot be vi-sualized using ultrasound.69 Occasionally, if the mass is largeenough to be visualized with ultrasound, a biopsy can be per-formed.69 This is done using a coaxial technique and anintercostal or subcostal approach, with the biopsy side up andthe patient in the decubitus position.

In a study of biopsies of 50 indeterminate adrenalmasses performed using ultrasound or CT guidance byPaulsen et al,69 a correct diagnosis was made in 96%. Theseincluded metastatic disease (70%), adrenocortical cancer (8%),adenoma (12%), and pheochromocytoma (6%). Some authorsdo not recommend biopsy of a possible pheochromocytomabecause of the risk of severe hypertension.70 If this is the sus-pected diagnosis, this can usually be determined by clinicaland laboratory methods.

PERITONEAL AND MESENTERIC BIOPSYThis is performed on mesenteric masses (intra-abdominal

carcinomatosis, carcinoid) and lymph nodes (Table 6). Weuse a coaxial needle technique and check for blood vessellocation using color flow Doppler before advancing the nee-dle. Because most of these biopsies are performed based onCT findings, we make sure to note the position of loops ofbowel in relation to the mass (Fig. 9 and see Video, SDC 11,http://links.lww.com/RUQ/A61, which demonstrates a core bi-opsy of an omental mass obtained through a small amount ofascites in a patient with intraperitoneal carcinomatosis). Onemay traverse small bowel as its contents are sterile. We do notlike to go through the colon; however, other authors have doneso without complications. When loops of bowel are in the way,they can be displaced by putting pressure on the abdominalwall with the transducer.71 Once the coaxial needle tip is at the

FIGURE 8. Abdominal CT scan of the abdomen withoutcontrast in a patient who developed a hematoma after renalbiopsy in the right retroperitoneum (arrows).

TABLE 6. Indication for Mesenteric/Omental/LymphNode Biopsy

Peritoneal carcinomatosisVovarian carcinoma, gastrointestinal malignancies(pancreatic, gastric, or colon cancer)

LymphomaVperitoneal lymphomatosis in non-Hodgkin disease

Carcinoid

DesmoidVGardner syndrome

Lymphadenopathy

Primary peritoneal mesothelioma

Amyloidosis




biopsy site, the throw of the cutting needle should be adjustedso that the well of the needle will end up within the mass. If itextends too far beyond the mass, it may lacerate an adjacentmesenteric vessel leading to bleeding. If lymphoma is thesuspected diagnosis, we perform FNAs (Fig. 10, see Video,SDC 12, http://links.lww.com/RUQ/A62, which demonstratesFNAs of an omental implant in a patient with a history of lym-phoma) with or without a coaxial needle (see Video, SDC 13,http://links.lww.com/RUQ/A63, which demonstrates FNAsof iliac lymph nodes in a patient with a remote history ofbreast cancer).

In a study using both CT (82%) and ultrasound guidance(18%), Souza and his colleagues72 had a success rate of 89%on 111 patients with peritoneal/omental masses. The majorityof these masses were malignant, and in patients with a knownprimary, the biopsy revealed a second malignant tumor in10%. They used FNAs (58.6%), a combination of FNAs andcore needles (38.7%), and core needles only (3%). They had3 minor complications: pain requiring medication in 1 patient,hypotension requiring fluid resuscitation in another, and 1

patient had a small hematoma at the biopsy site. Another com-plication can be cellulitis in the abdominal wall at the needleentry site and rarely peritonitis.73 In a study by Ho et al,73 theymade a correct diagnosis in 23 (92%) of 25 omental masses,whereas Spencer et al74 had a lesser rate of diagnosis of 77%.

RETROPERITONEAL/PELVIC BIOPSYLymphadenopathy and masses in the retroperitoneum/

pelvis are often too small to have a biopsy performed on us-ing ultrasound unless the patient is thin enough to allow vi-sualization (Table 7). When they are sonographically visiblethrough the anterior abdominal with the patient supine, thisapproach can be used.75 Pressure with the transducer on theanterior abdominal wall will shorten the needle path.71 Al-though bowel may be traversed, careful attention should bepaid to avoid mesenteric vessels. For this reason, we prefer thedecubitus or prone approach through the back if this allowsvisualization of the lymph node/mass (Fig. 11). In the pelvis,the transgluteal route can be used, making sure to avoid thearea of the sciatic nerve.

We use a coaxial technique, and when lymphoma is sus-pected, we perform both FNAs and core biopsies as previouslydescribed. Using FNAs on retroperitoneal lymph nodes (meandiameter, 2.1 cm), Fisher et al71 had a success rate of 86%.They had 1 major complication due to injury to the inferiorepigastric artery that required embolization (Fig. 12). Yarramet al76 evaluated the effectiveness of ultrasound (60%) andCT-guided (40%) core biopsies in diagnosing pelvic masses.The majority of the masses were malignant (70%), and ultra-sound guidance had an accuracy of 95.4% compared with84.6% for CT. In a study by Memel et al,75 they had a successrate of 91% when the lymph nodes could be visualized forbiopsy.

ABDOMINAL WALL BIOPSYMasses of the abdominal wall include primary mes-

enchymal mass such as a desmoid tumor and metastaticlesions (Table 8). These biopsies are performed using a high-frequency transducer and a freehand coaxial technique. Thethrow of the needle needs to be adjusted to reflect the size ofthe mass. The coaxial needle is brought in from the side so thatthe biopsy will be along the longest axis of the mass. This tech-nique is both safe and effective. If lymphoma is the suspecteddiagnosis, we perform FNAs using the technique describedpreviously (see Video, SDC 14, http://links.lww.com/RUQ/A65,

FIGURE 9. Contrast-enhanced CT scan of the abdomenshowing ascites (asterisk) with enhancement and thickeningof the peritoneum as well as irregular thickening of theomentum (arrows) in a patient with peritoneal carcinomatosis.

FIGURE 10. Contrast-enhanced CT scan of the abdomenshowing a well-defined solid omental mass (arrow).

TABLE 7. Indication for Retroperitoneal/Pelvic Biopsy

Lymphoma

LymphadenopathyVmetastatic

Retroperitoneal fibrosis

Primary malignancies

Liposarcoma

Malignant fibrous histiocytoma

Leiomyosarcoma

Benign tumors

Paraganglioma

Neurofibroma




which demonstrates FNAs in an anterior abdominal wallmetastatic lesion of a patient with squamous cell carcinoma ofthe tongue).

POSTBIOPSY OBSERVATIONFollowing the biopsy, we assess patients for pain, and we

obtain a baseline postprocedure set of vital signs in the ul-trasound department. Those who had a core biopsy of a solidorgan are then transferred to the postanesthesia care unit forat least 2 hours of observation in the supine position at com-plete bed rest. For patients undergoing renal biopsies, differentauthors recommend varied observation periods ranging frompostprocedure hospitalization with 24 hours of rest after per-cutaneous renal biopsy as 33% of complications occur after8 hours55 to releasing patients after only 6 or 8 hours of ob-servation.59,77 At our institution, when renal biopsies are per-formed with the nephrologists, patients are kept in the hospitalovernight for an observation period, which totals approxi-mately 18 hours, and this is usually because of associatedcomorbidities. So far, our nephrologists have not reported any

major postbiopsy complications. Others advocate that dis-charging asymptomatic patients 1 hour after liver biopsies per-formed with an 18-gauge needle is safe and cost-effective.78,79

Their pulse rate, blood pressure, and O2 saturation are moni-tored. We do not remeasure the hemoglobin and hematocrit.Some patients experience pain following the biopsy, which wetreat with acetaminophen 650 mg given orally. Uncommonly,we have to use oxycodone orally or injectable narcotic drugssuch as morphine sulfate intramuscularly to control the pain.This may prolong their stay in the postanesthesia care unit.Patients who have core biopsies of lymph nodes or FNAs of asolid organ, mass, or lymph node are discharged immediatelyfollowing the procedure, provided they are asymptomatic.

Before being discharged, patients are instructed onwound care and on resumption of their medications that mayhave been stopped before the procedure. They are also advisedto avoid vigorous activity including heavy lifting or pushingfor at least 1 week. Patients are given a 24-hour number to callone of our on-call physicians in case they experience weak-ness, dizziness, or any new pain after going home.

The pathology report is usually available within a week.Patients are instructed to call their doctor’s office at that timeto make an appointment to discuss the results. We review allof our pathology reports and tabulate the results to determineour success rate in making a pathological diagnosis. Any com-plications are also discussed at our regular quality assurancemeetings.

FIGURE 11. A, Contrast-enhanced CT scan of the abdomen showing a large well-defined hypodense solid mass in theretroperitoneum (asterisk) displacing the pancreas anteriorly. B, Ultrasound-guided biopsy of the retroperitoneal mass showingthe biopsy needle tip within the mass (arrow).

FIGURE 12. Contrast-enhanced CT scan of the abdomen inthe same patient as in Figure 11, who developed pancreatitisafter biopsy of the large retroperitoneal mass. There is apseudocyst anterior to the pancreas (arrow).

TABLE 8. Indication for Anterior Abdominal Wall Biopsy

Benign tumors

Neurogenic tumorsVschwannomas, neurofibromas

Desmoid tumors

Malignant tumors

Malignant fibrous histiocytoma

Lymphoma

Liposarcomas

Rhabdomyosarcoma

Melanoma

MetastaticVSister Mary Joseph nodule




CONCLUSIONSUltrasound is a safe, effective, and convenient method

for performing image-guided biopsies, often preferred as it isrelatively inexpensive and portable and does not depend onionizing radiation. It is, however, operator dependent, and itsuse is limited to sonographically visible lesions. We review theapproach to common targets of ultrasound-guided biopsy, withspecific emphasis on procedure safety and efficiency. As post-biopsy hemorrhage is the most worrisome complication, wediscuss our approach to the patient’s coagulation profile, in-cluding routine screening with recent complete blood count,PT, INR/aPTT, and review of coagulation-affecting medica-tions before the procedure. It is emphasized that this is per-formed on an individual basis, with consideration to thepatient’s underlying pathology and appropriate and timelycoordination with the ordering physicians. We address detailsof patient preparation, needle and approach selection, samplepreparation with a view toward suspected pathology, and rec-ommendations for postbiopsy observation and discharge in-structions. Finally, we stress the importance of regular reviewof pathology reports and tabulation of results to improve futurepractice.

REFERENCES1. Reading CC, Charboneau JW, James EM, et al. Sonographically guided

percutaneous biopsy of small (3 cm or less) masses. AJR Am JRoentgenol. 1988;151:189Y192.

2. NolsLe C, Nielsen L, Torp-Pedersen S, et al. Major complications anddeaths due to interventional ultrasonography: a review of 8000 cases.J Clin Ultrasound. 1990;18:179Y184.

3. Yu SC, Liew CT, Lau WY, et al. US-guide percutaneous biopsy of small(G or = 1-cm) hepatic lesions. Radiology. 2001;218(1):195Y199.

4. Kliewer MA, Sheafor DH, Paulson EK, et al. Percutaneous liver biopsy: acost-benefit analysis comparing sonographic and CT guidance.AJR Am J Roentgenol. 1999;173:1199Y1202.

5. Malloy PC, Grassi CJ, Kundu S, et al. Consensus guidelines forperiprocedural management of coagulation status and hemostasis riskin percutaneous image-guided interventions. J Vasc Interv Radiol.2009:20:S240YS249.

6. O’Connor SD, Taylor AJ, Williams EC, et al. Coagulation conceptsupdate. AJR Am J Roentgenol. 2009;193:1656Y1664.

7. Atwell TD, Smith RL, Hesley GK, et al. Incidence of bleeding after15,181 percutaneous biopsies and the role of Aspirin. AJR Am JRoentgenol. 2010;194:784Y789.

8. Caturelli C, Squillante MM, Andriulli A, et al. Fine-needle liver biopsy inpatients with severely impaired coagulation. Liver. 1993;13:270Y273.

9. Winter TC, Lee FT, Hinshaw JL. Ultrasound-guided biopsies in theabdomen and pelvis. Ultrasound Q. 2008;24:45Y68.

10. Sibon I, Orgogozo JM. Antiplatelet drug discontinuation is a risk factorfor ischemic stroke. Neurology. 2004;62:1187Y1189.

11. Maulaz AB, Bezerra DC, Michael P, et al. Effect of discontinuing aspirintherapy on the risk of brain ischemic stroke. Arch Neurol.2005;62:1217Y1220.

12. Lecompte T, Hardy JF. Antiplatelet agents and perioperative bleeding.Can J Anaesth. 2006;53:S103YS112.

13. Youssef WI, Salazar F, Dasarathy S, et al. Role of fresh frozen plasmainfusion in correction of coagulopathy of chronic liver disease: adual-phase study. Am J Gastroenterol. 2003;98(6):1391Y1394.

14. Abdel-Wahab OI, Healy B, Dzik WH. Effects of fresh-frozen plasmatransfusion on prothrombin time and bleeding in patients with mildcoagulation abnormalities. Transfusion. 2006;46(8):1279Y1285.

15. Caturelli E, Giacobbe A, Facciorusso D, et al. Percutaneous biopsy indiffuse liver disease: increasing diagnostic yield and decreasingcomplication rate by routine ultrasound assessment of puncture site.Am J Gastroenterol. 1996;91:1318Y1321.

16. Mueller PR, Biswal S, Halpern EF, et al. Interventional radiologic

procedures: patient anxiety, perception of pain, understandingof procedure, and satisfaction with medicationVa prospective study.Radiology. 2000;215:684Y688.

17. Sistla SC, Prabhu G, Sistla S, et al. Minimizing wound contaminationin a ‘‘clean’’ surgery: comparison of chlorhexidine-ethanol andpovidone-iodine. Chemotherapy. 2010;56:261Y267.

18. Noorani A, Rabey N, Walsh SR, et al. Systematic review andmeta-analysis of preoperative antisepsis with chlorhexidine versuspovidone-iodine in clean-contaminated surgery. Br J Surg.2010;97:1614Y1620.

19. Darouiche RO, Wall MJ, Itani KM, et al. Chlorhexidine-alcohol versuspovidone-iodine for surgical-site antisepsis. N Engl J Med.2010;362:18Y26.

20. Feld R, Needleman L, Goldberg BB. Use of needle-vibrating device andcolor Doppler imaging for sonographically guided invasive procedures.AJR Am J Roentgenol. 1997;168:255Y256.

21. Bisceglia M, Matalon TA, Silver B. The pump maneuver: an atraumaticadjunct to enhance US needle tip localization. Radiology.1990;176:867Y868.

22. Hamper UM, Svader BL, Sheth S. Improved needle-tip visualizationby color Doppler sonography. AJR Am J Roentgenol.1990;156:401Y402.

23. Hopper KD, Abendroth CS, Sturtz KW, et al. Blinded comparison ofbiopsy needles and automated devices in vitro: 1. Biopsy of diffusehepatic disease. AJR Am J Roentgenol. 1993;161:1293Y1297.

24. Hopper KD, Abendroth CS, Sturtz KW, et al. Blinded comparison ofbiopsy needles and automated devices in vitro: 2. Biopsy of diffusehepatic disease. AJR Am J Roentgenol. 1993;161:1299Y1301.

25. Jennings PE, Donald JJ, Coral A, et al. Ultrasound-guided core biopsy.Lancet. 1989;1:1369Y1371.

26. Arabas KB, Dingil G, Ungul U, et al. Accuracy and safety of percutaneousUS-guided needle biopsies liver metastasis and hemangiomas.Minerva Gastroenterol Dietol. 2010;56:377Y382.

27. Hopper KD, Abendroth CS, TenHave TR, et al. Multiple fine-needlebiopsies using a coaxial technique: efficacy and a comparison ofthree methods. Cardiovasc Intervent Radiol. 1995;18:307Y311.

28. Rizvi SA, Husain M, Khan S, et al. A comparative study of fine needleaspiration cytology versus non-aspiration technique in thyroid lesions.Surgeon. 2005;3:273Y276.

29. Fagelman D, Chess Q. Non-aspiration fine-needle cytology of the liver: anew technique for obtaining diagnostic samples. AJR Am J Roentgenol.1990;155:1217Y1219.

30. Weigand K. Percutaneous liver biopsy: retrospective study over15 years comparing 287 inpatients with 428 outpatients.J Gastroenterol Hepatol. 2009;792Y799.

31. Ishikawa T, Kamimura H, Tsuchiya A, et al. Comparison of a newaspiration needle device and the Quick-Core biopsy for transjugular liverbiopsy. World J Gastroenterol. 2006;12:6339Y6342.

32. Gorriz E, Reyes R, Lobrano MB, et al. Transjugular liver biopsy: a reviewof 77 biopsies using a spring-propelled cutting needle (biopsy gun).Cadiovasc Intervent Radiol. 1996;19:442Y445.

33. Maturen KE, Nghiem HV, Marrero JA, et al. Lack of tumor seeding ofhepatocellular carcinoma after percutaneous needle biopsy using coaxialcutting needle technique. AJR Am J Roentgenol. 2006;187:1184Y1187.

34. Hatfield MK, Beres RA, Sane SS, et al. Percutaneous imaging-guidedsolid organ core needle biopsy: axial versus nonaxial method.AJR Am J Roentgenol. 2008;190:413Y417.

35. Sparchez Z. Complications after percutaneous liver biopsy in diffusehepatopathies. Rom J Gastroenterol. 2002;11:335Y341.

36. Erturk SM, Mortele KJ, Tuncali K, et al. Fine-needle aspiration of solidpancreatic masses: comparison of CT and endoscopic sonographyguidance. AJR Am J Roentgenol. 2006;187:1531Y1535.

37. Tseng HS, Chen CY, Chan WP, et al. Percutaneous transgastric computedtomographyYguided biopsy of the pancreas using large needles.World J Gastroenterol. 2009;15:5972Y5975.

38. Brandt KR, Charboneau WJ, Stephens DH, et al. CT- and US-guidedbiopsy of the pancreas. Radiology. 1993;187(1):99Y104.

39. Paulsen SD, Nghiem HV, Negussie E, et al. Evaluation of imaging-guidedcore biopsy of pancreatic masses. AJR Am J Roentgenol. 2006;187:769Y772.

40. Matsubara J, Okusaka T, Morizane C, et al. Ultrasound-guidedpercutaneous pancreatic tumor biopsy in pancreatic cancer: a comparison




with metastatic liver tumor biopsy, including sensitivity, specificity,and complications. J Gastroenterol. 2008;43:225Y232.

41. DelMaschio A, Vanzulli A, Sironi S, et al. Pancreatic cancer versuschronic pancreatitis: diagnosis with CA 19-9 assessment, US, CTand CT-guided fine-needle biopsy. Radiology. 1991;178:95Y99.

42. Elvin A, Anderson T, Scheibenpflug L, et al. Biopsy of the pancreaswith a biopsy gun. Radiology. 1990;176:615Y616.

43. Karlson BM, Forsman CA, Wilander E, et al. Efficiency of percutaneouscore biopsy in pancreatic tumor diagnosis. Surgery. 1996;120:75Y79.

44. Di Stasi M, Lencioni R, Solmi L, et al. Ultrasound-guided fine needlebiopsy of pancreatic masses: results of a multicenter study.Am J Gastroenterol. 1998;93:1329Y1333.

45. Zamboni GA, D’Onofrio M, Idili A, et al. Ultrasound-guidedpercutaneous fine-needle aspiration of 545 focal pancreatic lesions.AJR Am J Roentgenol. 2009;193:1691Y1695.

46. Zech CJ, Helmberger T, Wichmann MW, et al. Large core biopsy of thepancreas under CT fluoroscopy control: results and complications.J Comput Assist Tomogr. 2002;26:743Y749.

47. Bergenfeldt M, Genell S, Lindholm K, et al. Needle-tract seeding afterpercutaneous fine-needle biopsy of pancreatic carcinoma. Case report.Acta Chir Scand. 1988;154:77Y79.

48. de Sio I, Castellano L, Clandra M, et al. Subcutaneous needle-tractseeding after fine needle aspiration biopsy of pancreatic liver metastasis.Eur J Ultrasound. 2002;15:65Y68.

49. Kosugi C, Furuse J, Ishii H, et al. Needle tract implantation ofhepatocellular carcinoma and pancreatic carcinoma afterultrasound-guided percutaneous puncture: clinical and pathologiccharacteristics and the treatment of needle tract implantation.World J Surg. 2004;28:29Y32.

50. Keogan MT, Freed KS, Paulson EK, et al. Imaging-guided percutaneousbiopsy of focal splenic lesions: update on safety and effectiveness.1999;172:933Y937.

51. Lieberman S, Libson E, Sella T, et al. Percutaneous image-guided splenicprocedures: update on indications, technique, complications, andoutcomes. Semin Ultrasound CT MR. 2007;28:57Y63.

52. McInnes MD, Kielar AZ, Macdonald DB. Percutaneous image-guidedbiopsy of the spleen: systematic review and meta-analysis of thecomplication rate and diagnostic accuracy. Radiology. 2011;260(3):699Y708.

53. Lucey BC, Boland GW, Maher MM, et al. Percutaneous nonvascularsplenic intervention: a 10-year review. AJR Am J Roentgenol.2002;179:1591Y1596.

54. Kang M, Kalra N, Gulati M, et al. Image guided percutaneous splenicinterventions. Eur J Radiol. 2007;64:140Y146.

55. Whittier WL, Korbet SM. Renal biopsy: update. Curr Opin NephrolHypertens. 2004;13:661Y665.

56. Lebret T, Poulain JE, Molinie V, et al. Percutaneous core biopsy forrenal masses: indications, accuracy and results. J Urol.2007;178:1184Y1188.

57. Caoili EM, Bude RO, Higgins EJ, et al. Evaluation of sonographicallyguided percutaneous core biopsy of renal masses. AJR Am J Roentgenol.2002;179:373Y378.

58. Constantin A, brisson ML, Kwan J, et al. Percutaneous US-guidedrenal biopsy: a retrospective study comparing the 16-gauge end-cutand 14-gauge side-notch needles. J Vasc Interv Radiol.2010;21:357Y361.

59. Maya ID, Allon M. Percutaneous renal biopsy: outpatient observationwithout hospitalization is safe. Semin Dial. 2009;22:458Y461.

60. Johnson PT, Nazarian LN, Feld RI, et al. Sonographically-guidedrenal mass biopsy: indications and efficacy. Ultrasound Med.2001;749Y753.

61. Song JH, Chaudry FS, Mayo-Smith WW. The incidental adrenal masson CT: prevalence of adrenal disease in 1049 consecutive adrenal massesin patients with known malignancy. AJR Am J Roentgenol. 2008;190:1163Y1168.

62. Boland GW. Adrenal imaging: why, when, what, and how? Part 1. Whyand when to image. AJR Am J Roentgenol. 2010;195:W377YW381.

63. Boland GW. Adrenal imaging: why, when, what, and how? Part 2. Whattechnique? AJR Am J Roentgenol. 2011;191:W1YW5.

64. Boland GW. Adrenal imaging: why, when, what, and how? Part 3. Thealgorithmic approach to definitive characterization of the adrenalincidentaloma. AJR Am J Roentgenol. 2011;195:W109YW111.

65. Berland LL, Silverman SG, Gore RM, et al. Managing incidentalfindings on abdominal CT: white paper of the ACR incidental findingscommittee. J Am Coll Radiol. 2010;7:754Y773.

66. Young WF. The incidentally discovered adrenal mass. N Engl J Med.2007:356Y601.

67. Casola G, Nicolet V, vanSonnenberg E, et al. Unsuspectedpheochromocytoma: risk of blood-pressure alteration during percutaneousadrenal biopsy. Radiology. 1986;159:733Y735.

68. McCorkell SJ, Nilesa NL. Fine needle aspiration ofcatecholamine-producing adrenal masses: a possible fatal mistake.AJR Am J Roentgenol. 1985;145:113Y114.

69. Paulsen SD, Nghiem HV, Korobkin M, et al. Changing role ofimaging-guided percutaneous biopsy of adrenal masses: evaluationof 50 adrenal biopsies. AJR Am J Roentgenol. 2004;182(4):1033Y1037.

70. Vanderveen KA, Thompson SM, Callstrom MR, et al. Biopsy ofpheochromocytomas and paragangliomas: potential for disaster.Surgery. 2009;146:1158Y1166.

71. Fisher AJ, Paulson EK, Sheafor DH, et al. Small lymph nodes of theabdomen, pelvis, and retroperitoneum: usefulness of sonographicallyguided biopsy. Radiology. 1997;205:185Y190.

72. Souza FF, Mortele KJ, Cibas ES, et al. Predictive value of percutaneousimaging-guided biopsy of peritoneal and omental masses: results in111 patients. AJR Am J Roentgenol. 2009;192:131Y136.

73. Ho LM, Thomas J, Fine SA, et al. Usefulness of sonographic guidanceduring percutaneous biopsy of mesenteric masses. AJR Am J Roentgenol.2003;180:1563Y1566.

74. Spencer JA, Weston MJ, Saidi SA, et al. Clinical utility of image-guidedperitoneal and omental biopsy. Nat Rev Clin Oncol. 2010;7:623Y631.

75. Memel DS, Dodd GD, Esola CC. Efficacy of sonography as a guidancetechnique for biopsy of abdominal, pelvic, and retroperitoneal lymphnodes. AJR Am J Roentgenol. 1996;167:957Y962.

76. Yarram SG, Nghiem HV, Higgins E, et al. Evaluation of imaging-guidedcore biopsy of pelvic masses. AJR Am J Roentgenol. 2007;188:1208Y1211.

77. Lin WC, Yang Y, Wen YK, et al. Outpatient versus inpatient renal biopsy:a retrospective study. Clin Nephrol. 2006;66:17Y24.

78. Bicknell SG, Richenberg J, Cooperberg PL, et al. Early discharge aftercore liver biopsy: is it safe and cost-effective? Can Assoc Radiol J.2002;53:205Y209.

79. Pokorny CS, Waterland M. Short-stay, out-of-hospital, radiologicallyguided liver biopsy. MJA. 2002;176:67Y69.

APPENDIX

Video Clip 1Video that demonstrates deep injection of Lidocaine

to the level of the biopsy site (video; Supplementary DigitalContent 1, http://links.lww.com/RUQ/A48).

Video Clip 2Video that demonstrates the biopsy needle entering and

exiting the liver (video; Supplementary Digital Content 2,http://links.lww.com/RUQ/A49).

Video Clip 3Video that demonstrates FNAs of a small focal liver

lesion seen in a patient with a suspected diagnosis oflymphoma (video; Supplementary Digital Content 3,http://links.lww.com/RUQ/A50).

Video Clip 4Video that demonstrates a biopsy of the left lobe of

the liver in a patient with sarcoidosis using a shortthrow technique (video; Supplementary Digital Content 4,http://links.lww.com/RUQ/A51).




Video Clip 5Video that demonstrates the technique used for a

core biopsy from the left lobe of the liver, using a deepthrow technique (video; Supplementary Digital Content 5,http://links.lww.com/RUQ/A52).

Video Clip 6Video that demonstrates FNAs of a small focal

liver lesion (video; Supplementary Digital Content 6,http://links.lww.com/RUQ/A53).

Video Clip 7Video that demonstrates subsequent core biopsy of the

same liver lesion as in supplement digital content 6 (video; Sup-plementary Digital Content 7, http://links.lww.com/RUQ/A54).

Video Clip 8Video that demonstrates a core biopsy performed

on a deep liver lesion in a patient with a cirrhotic liverand ascites (video; Supplementary Digital Content 8,http://links.lww.com/RUQ/A55).

Video Clip 9Video that demonstrates core biopsy of the lower pole

of the kidney in a patient with IGA nephropathy (video; Sup-plementary Digital Content 9, http://links.lww.com/RUQ/A56).

Video Clip 10Video that demonstrates a core biopsy performed

of a large left lower pole renal mass in a patient withlymphoma (video; Supplementary Digital Content 10,http://links.lww.com/RUQ/A60).

Video Clip 11Video that demonstrates a core biopsy of an omental

mass obtained through a small amount of ascites in a patientwith intraperitoneal carcinomatosis (video; SupplementaryDigital Content 11, http://links.lww.com/RUQ/A61).

Video Clip 12Video that demonstrates FNAs of an omental implant in

a patient with a history of lymphoma (video; SupplementaryDigital Content 12, http://links.lww.com/RUQ/A62).

Video Clip 13Video that demonstrates FNAs of iliac lymph nodes in a

patient with a remote history of breast cancer (video; Supple-mentary Digital Content 13, http://links.lww.com/RUQ/A63).

Video Clip 14Video that demonstrates FNAs of an anterior abdomi-

nal wall metastatic lesion in a patient with squamous cellcarcinoma of the tongue (video; Supplementary Digital Con-tent 14, http://links.lww.com/RUQ/A65).




ultrasound-guided biopsies of the abdomen · pdf fileultrasound-guided biopsies of the abdomen...

Documents