Accuracy and Cost-Effectiveness of Core Needle Biopsy in the Evaluation of Suspected Lymphoma: A Study of 101 Cases

By Lachar, Whitney A Shahab, Imran; Saad, A Joe

* Context.-Lymphomas have traditionally been diagnosed on excisional biopsies of lymph nodes in order to evaluate tissue architecture and cytomorphology. Recent lymphoma classification schemes emphasize immunophenotypic, genetic, and molecular aspects in addition to morphology as diagnostic features. Core needle biopsies are increasingly being used to obtain tissue for diagnosis in patients with lymphadenopathy and a clinical suspicion of lymphoma. These procedures are rapid, minimally invasive, well tolerated, and may provide some architectural framework (unlike fine- needle aspirations), as well as material for ancillary studies. Objective.-To explore the accuracy, utility, and costeffectiveness of this technique.

Design.-Core needle biopsies of 101 consecutive patients from 2 large community hospitals who were suspected of having primary or recurrent lymphomas were retrospectively reviewed. All patients had hematoxylin-eosin- stained sections of needle cores. Specimens morphologically suspicious for lymphoma were subjected to ancillary studies, including immunohistochemistry, flow cytometry, and/or molecular studies. Core needle biopsy diagnoses were correlated with subsequent excisional biopsies, if performed.

Results.-Core needle biopsies established a definitive pathologic diagnosis for the vast majority of cases. A diagnosis was considered sufficient to begin treatment for primary and recurrent lymphomas in most cases. Compared with an open biopsy, there is a cost savings of greater than 75%.

Conclusion.-The accuracy of this technique, along with the cost savings and decreased morbidity, suggest that this method may be used safely and reliably as a first-line diagnostic technique.

(Arch Pathol Lab Med. 2007;131:1033-1039)

Core needle biopsy (CNB) is a technique that has been used to obtain tissue for diagnostic purposes for many years. More recently, the diagnosis of lymphoma has been attempted on CNBs, using ancillary studies to aid in the diagnosis and classification. Previously, classification schemes for lymphoma relied only on histomorphology, as in the Rappaport classification,1 which was developed in 1966. Subsequent classification schemes began to incorporate the immunophenotype of the malignant cells, but it was not until the Revised European American Lymphoma (REAL) classification that immunophenotype played a major role in the categorization of lymphomas.2 TheWorld Health Organization (WHO) modification of this scheme further emphasized the role of immunophenotypic data in the classification of lymphomas.3

Immunohistochemistry (IHC), flow cytometry, and molecular studies are increasingly relied upon to classify lymphoproliferative disorders, even in excisional biopsies, and can also often be used when only small amounts of tissue are obtained, as with CNB. In the past, deep lymph nodes and solid organ lesions could only be reached for biopsy by laparotomy or thoracotomy. Now, with the use of radiographically guided CNB, the morbidity and expense of these invasive procedures can be avoided. Core biopsies can be allocated for flow cytometry, molecular studies, and for IHC stains on paraffin-embedded tissue, in addition to hematoxylin-eosin sections, thus enabling the pathologist to not only diagnose but also subclassify the lymphoproliferative disorder with an accuracy that approaches that of an excisional biopsy.

This retrospective study comprised 101 patients with a clinical suspicion of lymphoma who underwent CNB for diagnosis at two major urban community hospitals. The goal of this study was to determine the accuracy and costeffectiveness of CNB in the initial and recurrent diagnosis and classification of lymphoma.

MATERIALS AND METHODS

We performed a retrospective analysis of all patients with lymphadenopathy and a clinical suspicion of lymphoma who underwent CNB at Methodist Dallas Medical Center and Presbyterian Hospital of Dallas in Dallas, Tex, between January 2002 and October 2005. Pathology records were reviewed to identify patients with lymphoid proliferations, both benign and malignant; nonlymphoid malignancies were excluded from the study. The data obtained included age; sex; site of biopsy; IHC, molecular study, and flow cytometry results; diagnosis; and follow-up biopsy results, if available.

Core biopsies were performed by a radiologist, generally using an 18-gauge cutting needle with or without additional passes using a 22- gauge aspiration needle. Biopsies were taken from various sites, including the retroperitoneum; superficial cervical, inguinal, or axillary lymph nodes; mediastinum; abdominopelvic lymph nodes; liver; spleen; kidney; lung; bone; and skin. All deep biopsies were radiographically guided, most using computed tomography imaging, and ultrasound was used occasionally. Each case had 2 to 4 passes with the cutting needle to obtain adequate material for histology and indicated ancillary studies. In one institution, if there was a clinical suspicion of lymphoma the radiologist sent 2 core biopsies to flow cytometry directly, without immediate assessment (20 cases). In the other institution, immediate interpretation was performed on Diff-quik-stained (Anapath, Lewisville, Tex) touch preparations of the core biopsy; if flow cytometry was indicated, 2 additional fine- needle aspiration (FNA) passes were made, rinsed in RPMI, and sent to the flow cytometry laboratory (12 cases).

Specimens for histology were fixed in 10% buffered formalin and were embedded in paraffin. Three levels, each 4 [mu]m thick, were stained with hematoxylin-eosin. Additional stains, including IHC and special stains for acid-fast bacilli (Kinyoun, Poly Scientific, Bay Shore, NY) and fungi (Grocott methenamine silver, Poly Scientific), were performed depending on hematoxylin-eosin morphologic findings. Immunohistochemistry analysis was performed on paraffin-fixed tissue according to the avidin-biotinperoxidase complex method using a panel of antibodies (Dako, Carpinteria, Calif) listed in Table 1.4 Flow cytometry was performed using a 4-color FACSCaliber cytometer (Becton Dickinson, San Diego, Calif), based on morphologic findings and clinical suspicion, as described above (Table 1). Flow cytometric studies were done on cell suspensions obtained from core biopsies or aspirated material performed at the same time. Immunohistochemistry and flow cytometry panels were tailored according to the morphologic impression and the amount of material available. In select cases, paraffin blocks were sent to Baylor University Medical Center (Dallas, Tex) for T-cell and B-cell gene rearrangement studies. These were performed using commercial kits (Tcell and B-cell clonality assays, InVivoScribe Technologies, San Diego, Calif). DNA was extracted from formalin-fixed, paraffinembedded tissue with the QIAmp DNA Mini Kit (Qiagen, Valencia, Calif). The polymerase chain reaction was performed in a GeneAmp PCR System 9700 thermocycler (Applied Biosystems, Foster City, Calif). The amplification products were then resolved by capillary electrophoresis in an ABI 3100 Genetic Analyzer (Applied Biosystems).5

The hospital and professional charges and payments were reviewed to compare the cost of the procedures at one of the institutions. Data were obtained from the hospital, a major medical billing company, surgeons, anesthesiologists, and radiologists. The specific Current Procedural Terminology codes for which charges and actual payments were made included 38500 (biopsy or excision of lymph node[s]; open, superficial), 38505 (biopsy or excision of lymph node[s]; by needle, superficial), 38510 (biopsy or excision of lymph node[s]; open, deep cervical node[s], 38525 (biopsy or excision of lymph node[s]; open, deep axillary node[s]), 38564 (limited lymphadenectomy for staging; retroperitoneal), 39000 (mediastinotomy with exploration, drainage, removal of foreign body, or biopsy; cervical approach), 76942 (ultrasonic guidance for needle placement, imaging supervision, and interpretation), and 76360 (computed tomography guidance for needle placement, radiological supervision, and interpretation). Charges and payments for the professional component of pathology were not included in the cost analysis, since they are identical for CNB and open lymph node biopsy (88305), immunohistochemical stains (88342 per antibody), and flow cytometry (88182, 88184, 88185, 88187, 88188, and 88189, depending on the number of markers analyzed).

RESULTS

Data from 101 consecutive patients who fit the above criteria were reviewed. The age range was 16 to 93 years, and the male- female ratio was 1:1. Of all cases, 20 were benign and included granulomatous inflammation and reactive hyperplasia (Figure 1, A and B). Lymphoma was diagnosed in 73 cases (Table 2). Of these, 14 had a previous diagnosis of lymphoma, and the remainder were primary lymphomas (Table 3). There were 8 cases diagnosed as atypical.

Hematoxylin-eosin stains alone were used in 14 cases. These included 12 cases of benign findings, such as granulomas and reactive hyperplasia, and 2 with insufficient material for further studies, both of which were diagnosed as atypical. The remaining 87 cases had additional studies. Immunohistochemistry was used in 79 cases, including 6 cases subsequently diagnosed as benign, 67 as lymphoma, and 6 as atypical. Flow cytometry was performed in 32 cases, 23 of which also had IHC performed. Flow cytometry was contributory to the diagnosis in 25 cases, of which 20 were lymphomas and 5 were benign. Gene rearrangement studies were performed on 9 cases, all of which also had IHC, and 4 of the 9 also had flow cytometry performed. Molecular studies were contributory in 6 of these cases, of which 5 had lymphoma diagnoses and 1 was reactive hyperplasia. The 3 remaining cases were negative for monoclonality by gene rearrangement but were determined to be lymphomas by other methods. Of all of the cases, 36 were diagnosed as diffuse large B-cell lymphomas. These were characterized by large cells with vesicular nuclei, single to multiple prominent nucleoli, and frequent mitotic figures (Figure 2, A and B). The typical phenotype included positivity for CD19, CD20, CD79a, and CD45, and occasional positivity for CD10, Bcl-2, and Bcl-6. Immunohistochemistry was used in all 36 cases. In addition, 9 had flow cytometry and 2 had gene rearrangement studies.

The second most frequent subtype (13 cases) was follicular lymphoma. These could be graded definitively in 4 cases, all of which were grade 1. The remaining 9 cases were classified as low grade or grades 1 to 2, with a comment that a higher grade component could not be completely excluded due to the nature of the sampling. Because grade 3 follicular lymphomas are treated differently, the distinction between grades 1 and 2 and grade 3 is important clinically. Some of the high-grade follicular lymphomas would be included with the large B-cell lymphomas. Grading of follicular lymphoma is based on the average number of intrafollicular centroblasts per highpower field.6 Grade 1 would have 0 to 5 centroblasts; grade 2, 6 to 15; and grade 3, more than 15. Follicular lymphomas can have a follicular or diffuse pattern, which can be difficult to determine on a small biopsy. The phenotype included positivity for CD19, CD20, CD79a, CD10, and Bcl-6. Intrafollicular lymphocytes were positive for Bcl-2. The cells were negative for CD5, CD43, and CD23. Ten of our cases were diagnosed using IHC, 7 using flow cytometry alone or in addition to IHC, and 2 using molecular studies in addition to IHC (Figure 3, A and B).

The third most frequent subtype was small lymphocytic lymphoma, which was diagnosed in 6 cases. These were characterized by small, monotonous lymphocytes with rounded nuclei and scant cytoplasm. They were generally positive for CD19, CD20, CD79a, CD23, CD5, and CD43, and negative for CD10 and cyclin D1. A total of 3 of these cases were diagnosed by IHC alone, and 3 were diagnosed using flow cytometry alone.

Classical Hodgkin lymphoma was diagnosed in 5 cases, which were characterized by the presence of classic Reed- Sternberg (RS) cells or RS variants, lacunar cells, and a background of lymphocytes, eosinophils, neutrophils, and plasma cells. The RS cells were positive for CD30 in all cases, and were negative for CD15, CD45, CD3, CD5, and CD20. All 5 of these cases were diagnosed using IHC only (Figure 4, A and B).

There were 3 cases diagnosed as marginal zone lymphoma that were characterized by small to medium cells with indented nuclei and abundant cytoplasm. They were all diffuse proliferations. The phenotype included positivity for CD20, CD79a, and CD43, and negativity for CD5, CD10, CD23, cyclin D1, and Bcl-6. All 3 of these cases were diagnosed using IHC.

Mantle cell lymphoma was diagnosed in 2 cases. These were characterized by small to medium lymphocytes with round to irregular nuclei and scant cytoplasm. The cells were positive for CD19, CD20, CD79a, CD5, CD43, and cyclin D1, and they were negative for CD23 and CD10. Both of our cases were diagnosed using IHC, and one also used flow cytometry.

A single case of precursor B-lymphoblastic lymphoma was diagnosed. Neoplastic lymphocytes were mediumsize cells with irregular nuclei, scant cytoplasm, inconspicuous nucleoli, and frequent mitotic figures. The cells were positive for CD19, CD79a, CD43, TdT, CD34, CD20, and CD10. The case was diagnosed using both IHC and flow cytometry.

There was a single case diagnosed as anaplastic largecell lymphoma that had large pleomorphic cells with multiple prominent nucleoli and abundant cytoplasm. The pattern was diffuse, and it lacked the reactive background usually seen in Hodgkin lymphoma. Immunohistochemistry alone was used to make the diagnosis. The cells were positive for CD30 and CD4, and were negative for Alk-1 and CD138.

We had a single composite lymphoma in a CNB of the liver, which had both grade 3 follicular lymphoma and marginal zone lymphoma components. Gene rearrangement studies showed a biclonal population of B cells. Using IHC, all cells were CD5 negative and CD20 positive, a portion co-expressed CD43, and some were positive for Bcl-2. The larger cells were positive for Bcl-6 and CD10, whereas the smaller cells were negative for both.

Five cases were diagnosed as lymphomas but could not be further classified. In one case, the patient had a history of follicular lymphoma, grade 1; the current CNB had an abnormal morphology thought to be lymphoma, but the IHC results were inconclusive for classification. The second case was thought to be a grade 2 or 3 follicular lymphoma or a diffuse large B-cell lymphoma, but the small biopsy size precluded definitive diagnosis. The third case was a lymphoma that had a heterogeneous background cell population, including neutrophils and plasma cells, and there was a suspicion of Hodgkin lymphoma, but diagnostic cells were not present in the biopspy. In this case an excisional biopsy confirmed Hodgkin lymphoma. In the fourth case, the patient had a history of diffuse large B-cell lymphoma of the stomach, and the CNB of the retroperitoneal mass had predominantly small round lymphocytes, with inconclusive IHC results. Gene rearrangement studies showed that the cells represented a monoclonal B-cell population; thus, a diagnosis of B-cell lymphoma was made. The fifth case had extensive necrosis and rare atypical lymphocytes, was negative flow cytometry studies, and was positive for B-cell gene rearrangement by polymerase chain reaction (PCR). It could not be further classified due to a paucity of viable cells.

In most cases no follow-up material was obtained, because patients were treated based on the CNB diagonsis. Subsequent excisional biopsies were performed in only 5 cases. Two patients died and had autopsies performed. Both autopsies confirmed the mediastinal CNB diagnoses of marginal zone and diffuse large B-cell lymphomas.

The local charges for the biopsies varied widely depending on the site and payor; therefore, the numbers provided are a relative comparison. Hospital billing data were obtained for 27 CNB and 36 open lymph node biopsies of superficial sites for Medicare, HMO, PPO, and commercial insurance. On average, hospital charges and payments for a needle biopsy were 76.7% less than those for an open biopsy, with charges ranging from $1861 to $12 956 and payments from less than $500 to more than $10 000. Surgeons’ charges and payments could only be obtained for biopsies of superficial lymph nodes from both hospital and outpatient surgery centers. No data could be obtained from cardiothoracic surgeons. Too few retroperitoneal biopsies were performed to provide meaningful numbers. Surgeons’ charges for a superficial lymph node biopsy were similar to those of radiologists for a needle biopsy, but surgeons were paid on average 1.5 to 2 times as much for PPO/HMO patients and more than twice as much for Medicare patients. Anesthesiologists’ charges also added significantly to the cost of an open biopsy. Table 4 shows the average payments for CNB and superficial lymph node biopsy. Biopsies of mediastinal, retroperitoneal, and deep-seated lesions are not included due to insufficient data. The radiologists’ payments include imaging studies used to acquire the biopsy but not diagnostic imaging performed prior to the procedure. Also of note is the fact that Medicare does not reimburse for conscious sedation.

COMMENT

Due to the enhancements to and advances in immunophenotyping techniques that are now widely available, the ability to diagnose and subclassify lymphomas may require only a small amount of tissue. Nodal architecture is much less important in the most current classification systems; thus, a complete nodal excision often contributes no more to an accurate diagnosis than does a CNB. Because of this, CNBs are used with increasing frequency in patients with lymphadenopathy and a suspicion of lymphoma. Accuracy in obtaining adequate tissue in deep-seated lesions has also increased with the refinement of radiographically guided biopsy techniques and onsite evaluation for adequacy.7-12 In this study, the sensitivity for establishing a diagnosis of lymphoma was 90%. The sensitivity of lymphoma diagnoses from excisional biopsies of lymph nodes ranges from 80% to 100%; thus, it appears that diagnoses from CNB are as accurate as those from more invasive procedures.13 An unequivocal diagnosis was made in 91% of biopsies in this study, which approximates the rates found in other studies of 72% to 100%.7-9,11- 15 Although it has been argued that CNB is best used to diagnose recurrent disease, the correct classification rate of new diagnoses in this study is 95%, which demonstrates that this technique is equally useful for primary diagnoses.

Few studies on CNB for the diagnosis of lymphoma pertaining to diagnostic surgical pathology are reported in English-language literature. There are several reviews in the radiology literature that do not specifically address the histomorphologic evaluation of these increasingly encountered specimens. It is important for pathologists to understand the goal of an accurate diagnosis and classification based on a minimal amount of tissue, and to use all of the methods currently available to achieve this goal. Although core biopsies were previously used more as a screening technique prior to subsequent excisional biopsy, this is no longer the case; thus, every attempt should be made to obtain a definitive diagnosis from the needle cores. Fine-needle aspiration also has been used with some frequency as a minimally invasive technique to obtain diagnostic material. Some studies have shown that FNA can be accurate in the diagnosis and classification of lymphomas, with accuracy rates ranging from 12% to 82% and varying significantly depending on concurrent use of flow cytometry.16-18 A controversial study by Hehn et al19 that accounted for the lowest accuracy rate of 12% found that FNA did not give specific and complete diagnoses in most cases, and thus required subsequent excisional biopsy before beginning treatment. However, this study was composed of referred cases, very few of which had ancillary studies, and thus they relied purely on morphology for a diagnosis. Conclusions about the accuracy of FNA cannot be made from this study, since it does not incorporate immunology or genetics. Even so, the highest accuracy rate within published studies is 82%. Although this may be the biopsy method of choice in some circumstances, accurate diagnoses are made less often than they are with CNB, which offers the distinct advantages of having tissue architecture to better assess morphology, as well as material for IHC and other ancillary studies, while only using a slightly larger needle. Radiographic assistance is used with both techniques for deep lesions, and the complications of both are similar.20 Advocates for FNA cytology in the diagnosis of lymphoma emphasize the need to obtain an adequate number of cells to perform the necessary studies. In an editorial, Katz21 recommended obtaining a minimum of 10 million cells by FNA without suction to accurately diagnose and subclassify lymphomas, with these cells counted using a cell counter at the time of biopsy. Cell counters are rarely used in a community setting. This may account for a lack of adequate material for phenotyping by flow cytometry in some cases. One study by Ravinsky and Morales17 found that for the diagnosis of lymphoma, FNA in conjunction with flow cytometry was accurate in 82% of cases, but the combination of FNA and CNB was 93% accurate, further supporting the utility of CNB. This study did not examine the accuracy of CNB alone; thus, it is not known whether the combination of the two procedures is more accurate than a single CNB procedure.

Core needle biopsy offers a distinct advantage over FNA in that it obtains intact tissue. Even though the entire nodal architecture is not visualized, tissue from a CNB offers some architecture to aid in diagnosis. Tissue is also used for IHC studies, including prognostic indicators. Gene rearrangement studies are easily performed on paraffin- embedded tissue, and there is also archival tissue for any future studies. These factors are important, particularly in the diagnosis of lymphoma, as several modalities are used to make an accurate diagnosis.

Previous studies on diagnosis of lymphoma on CNB tissue have not included gene rearrangement data.Molecular studies were used in 9 (8.9%) of our cases, and they supported a lymphoma diagnosis in 5 of those cases. This is an additional tool that can be used to aid in the diagnosis of lymphoma from small tissue cores, particularly when IHC stains and morphology are inconclusive and when fresh tissue is not available for flow cytometry studies.

In the vast majority of cases in our study, treatment was initiated based on the diagnosis from the CNB. There were 5 cases that were diagnosed as lymphoma but could not be further classified due to inconclusive IHC findings, necrotic biopsy material, and lack of sufficient material for further studies. Five cases were subsequently biopsied. One case was nondiagnostic on CNB, and thus underwent an excisional biopsy and was diagnosed as a recurrent follicular lymphoma. Two cases called atypical on CNB were diagnosed as diffuse large B-cell lymphoma on excisional biopsy. One was diagnosed as an unclassified lymphoma and was later found to be Hodgkin. The fifth case was diagnosed as Hodgkin lymphoma on CNB and was excised because the patient was obtunded and had pericardial and pleural effusions, and thus the clinicians wanted to further rule out a high-grade non-Hodgkin lymphoma. The CNB diagnosis was confirmed on excisional biopsy. Two patients died, and postmortem examinations were performed. In both cases, the CNB diagnoses were confirmed, one as marginal zone lymphoma and the other as diffuse large B-cell lymphoma.

In our study a single patient underwent a subsequent CNB 5 months after the original CNB. The first CNB diagnosis was low-grade B- cell lymphoma, and the second CNB was diagnosed as a diffuse large B- cell lymphoma. The discrepancy may be due to interpretive error, sampling error within the node, or large-cell transformation from a low-grade lymphoma. Chemotherapy was modi- fied based on the second biopsy. All of the cases included in this study were hospital patients who were selected as being clinically suspicious for lymphoma. In addition, they were generally of an older population, with an average age of 61 years. This population and the clinical presentation-driven selection process accounts for the high rate of lymphoma diagnoses in this study. By comparison, an outpatient population referred for FNA of a lymph node had a benign diagnosis 85% to 90% of the time (A.J.S., unpublished data, 2006).

The average payment for performing an open biopsy of a lymph node is $3529, but this varies depending on biopsy site, imaging modality, and type of anesthesia. Excisional biopsies carry a risk of infection, bleeding, and wound healing problems. There is a small risk of mortality due to anesthesia. They require sutures and can leave a significant scar. A day of recuperation is usually required. On the other hand, CNB carries a very low risk of infection and bleeding, and complications from general anesthesia are usually not seen, since most are performed under local anesthesia with or without intravenous sedation. There is minimal scarring, and sutures are not required. It is generally an outpatient procedure and requires only a few hours of observation after the procedure. The average payment for CNB is $822. The savings in performing a CNB instead of an open biopsy are thus $2707, or 76.7%. The cost in this study was calculated based on charges for biopsies of superficial sites. Cost savings are thus likely to be much greater when deep, technically more difficult and potentially more risky sites are included.

In addition, several factors that affect cost are not examined in this study, including costs associated with potential postoperative complications and lost wages from days away from work. These additional factors would most likely further amplify the cost discrepancy, particularly with open biopsy of deep-seated lesions. Only a small percentage (5%) of patients in this study underwent subsequent open biopsy; thus, the cost savings are real in that 95% of patients were treated without incurring the expense of an additional procedure. Even if all patients with atypical and nondiagnostic diagnoses were to later undergo an excisional biopsy, this would still be a minority of cases, and overall cost savings would still be significant. Finally, the pathologists’ charges were not included in the cost analysis. The Current Procedural Terminology codes, charges, and payments are identical for both procedures. At the two institutions, the workup is similar for a needle biopsy or excisional biopsy in a suspected lymphoma.

The findings of this study were reproducible within two large, urban tertiary care hospitals, and they support the findings in other studies that minimally invasive procedures can lead to a definitive diagnosis in a majority of cases. Subsequent excisional biopsies were performed in only 5 cases, which demonstrates that clinicians will and do treat patients based on a definitive diagnosis from a CNB without requiring an excisional biopsy.

References

1. Rappaport H. Tumors of the Hematopoietic System. Washington, DC: Armed Forces Institute of Pathology; 1966. Atlas of Tumor Pathology; section III.

2. Harris NL, Jaffe ES, Stein H, et al. A revised European American classifi- cation of lymphoid neoplasms: a proposal from the international lymphoma study group. Blood. 1994;84:1361-1392.

3. Harris NL, Jaffe ES, Diebold J, et al. The World Health Organization classification of neoplastic diseases of the hematopoietic and lymphoid tissue: report of the clinical advisory committee meeting, Airlie House, Virginia, November 1997. Histopathology. 2000;36:69-87.

4. Hsu SM, Raine L, Fanger H. The use of avidin-biotin peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabeled antibody (PAP) procedures. J Histochem Cytochem. 1981;29:577-580.

5. Sanchez S, Holmes H, Katabi N, et al. Composite lymphocyte- rich Hodgkin lymphoma and peripheral T-cell lymphoma associated with Epstein-Barr virus: a case report and review of the literature. Arch Pathol Lab Med. 2006;130:107- 112.

6. Nathwani BN, Harris NL, Weisenburger D, et al. Follicular lymphoma. In: Jaffe ES, Harris NL, Stein H,Vardiman JW, eds. Pathology and Genetics ofTumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2001:162- 167. World Health Organization Classification of Tumours; vol 3.

7. Agid R, Sklair-Levy M, Bloom AI, et al. CT-guided biopsy with cutting-edge needle for the diagnosis of malignant lymphoma: experience of 267 biopsies. Clin Radiol. 2003;58:143-147.

8. Ben-Yehuda D, Polliack A, Okon E, et al. Image-guided core- needle biopsy in malignant lymphoma: experience with 100 patients that suggests the technique is reliable. J Clin Oncol. 1996;14:2431- 2434. 9. De Kerviler E, Guermazi A, Zagdanski AM, et al. Image- guided core-needle biopsy in patients with suspected or recurrent lymphomas. Cancer. 2000;89:647- 652.

10. Pappa VI, Hussain HK, Reznek RH, et al. Role of image-guided core-needle biopsy in the management of patients with lymphoma. J Clin Oncol. 1996; 14:2427-2430.

11. Zinzani PL, Colecchia A, Festi D, et al. Ultrasound-guided core-needle biopsy is effective in the initial diagnosis of lymphoma patients. Haematologica. 1998;83:989-992.

12. Sklair-Levy M, Polliack A, Shaham D, et al. CT-guided core- needle biopsy in the diagnosis of mediastinal lymphoma. Eur Radiol. 2000;10:714-718.

13. Demharter J, Muller P, Wagner T, Schlimok G, Haude K, Bohndorf K. Percutaneous core-needle biopsy of enlarged lymph nodes in the diagnosis and subclassification of malignant lymphomas. Eur Radiol. 2001;11:276-283.

14. Screaton NJ, Berman LH, Grant JW. Head and neck lymphadenopathy: evaluation with US-guided cutting-needle biopsy. Radiology. 2002;224:75-81.

15. Zinzani PL, Corneli G, Cancellieri A, et al. Core needle biopsy is effective in the initial diagnosis of mediastinal lymphoma. Haematologica. 1999;84:600- 603.

16. Mourad WA, Tulbah A, Shoukri M, et al. Primary diagnosis and REAL/ WHO classification of non-Hodgkin’s lymphoma by fine-needle aspiration: cytomorphologic and immunophenoypic approach. Diagn Cytopathol. 2003;28: 191-195.

17. Ravinsky E, Morales C. Diagnosis of lymphoma by image-guided needle biopsies: fine needle aspiration biopsy, core biopsy or both? Acta Cytol. 2005; 49:51-57.

18. Gong JZ, Snyder MJ, Lagoo AS, et al. Diagnostic impact of core-needle biopsy on fine-needle aspiration of non-Hodgkin lymphoma. Diagn Cytopathol. 2004;31:23-30.

19. Hehn ST, Grogan TM, Miller TP. Utility of fine-needle aspiration as a diagnostic technique in lymphoma. J Clin Oncol. 2004;22:3046-3052.

20. Watkinson AF, Adam A. Complications of abdominal and retroperitoneal biopsy. Radiology. 1991;178:493-496.

21. Katz RL. Modern approach to lymphoma diagnosis by fine- needle aspiration. Cancer Cytopathol. 2005;105:429-431.

Whitney A. Lachar, MD; Imran Shahab, MD; A. Joe Saad, MD

Accepted for publication January 15, 2007.

From the Department of Pathology, Baylor University Medical Center, Dallas, Tex (Dr Lachar); the Department of Pathology, Presbyterian Hospital of Dallas, Dallas, Tex (Dr Shahab); and the Department of Pathology, Methodist Dallas Medical Center, Dallas, Tex (Dr Saad).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: A. Joe Saad, MD, Department of Pathology, Methodist Dallas Medical Center, 1441 N Beckley Ave, Dallas, TX 75203 (e- mail: joesaad@mhd.com).

Copyright College of American Pathologists Jul 2007

(c) 2007 Archives of Pathology & Laboratory Medicine. Provided by ProQuest Information and Learning. All rights Reserved.