Secondary Logo

Journal Logo

Comparison of the Yield of Transthoracic Needle Aspiration and Core Needle Biopsy Between Pulmonologist and Radiologist in a Community Based Practice

Daya, Ibrahim Abou MD*; Abhyankar, Archana MD; Timmireddy, Aruna MD*; Blum, Steve PhD; Patel, Yashwant D. MD; Diaz-Fuentes, Gilda MD, FCCP*

doi: 10.1097/LBR.0b013e31817f97e8
Original Investigations

Background Transthoracic needle aspiration (TTNA) and core needle biopsy (CNB) are performed by a pulmonologist or an interventional radiologist (IR). Very few pulmonary fellowship programs offer training in TTNA/CBN. We compared the yield and complications of TTNA/CNB when performed by pulmonologist versus IR.

Method This was a retrospective analysis of patients that underwent TTNA/CNB from 2003 to 2007 at our institution. Procedures were performed by either, any of 5 pulmonary fellows under the supervision of any of 4 pulmonologists or by a single IR.

Results Thirty-five patients were included, 19 in the pulmonary and 16 in the IR group. There were no differences in yield (P=0.28) or complications (P=0.31) between the groups. The overall diagnostic yield was 63% for pulmonary versus 69% for IR. Malignancy was the most common diagnosis (70%). CNB was used by pulmonary in 21% compared with 50% of IR cases. The yield was higher when TTNA was combined with CBN (75% vs. 61% for TTNA alone). An on-site pathologist was present only during the pulmonary cases.

Conclusions The overall yield and complications were similar in both groups. Combining TTNA and CBN provides higher yield than TTNA alone. Obtaining expertise in CNB is of the outmost importance to maximize yield and decrease need for more invasive procedures. Pulmonary fellowship programs should continue to offer training in TTNA/CNB with an onsite pathologist where available, to achieve diagnostic yield comparable with the interventional radiologist. Those programs should develop a system to maintain proficiency for the faculty.

*Division of Pulmonary Medicine

Departments of Radiology

Medicine, Bronx Lebanon Hospital Center, New York

No financial support was used for the study and no reprints will be ordered.

Reprints: Gilda-Diaz Fuentes, MD, FCCP, Albert Einstein College of Medicine, Pulmonary and Critical Care, Bronx Lebanon Hospital Center, 1650 Grand Concourse, Pulmonary Division, Bronx, NY 10457 (e-mail:

Received for publication May 6, 2008; accepted May 12, 2008

There is no conflict of interest.

Back to Top | Article Outline


Transthoracic needle aspiration (TTNA) and biopsy is the percutaneous sampling of lesions involving the chest wall, lung parenchyma, and mediastinum for cytologic, histopathologic, or microbiologic examinations.

Common indications for TTNA/CNB include the diagnosis of lung nodules or infiltrates, pleural masses, mediastinal masses, and other thoracic lesions accessible through percutaneous approach.1

TTNA/CNB has an overall diagnostic sensitivity of 68% to 96%, a specificity of 100%, and an accuracy of 74% to 96% in lesions of all sizes. In the smaller lesions, its diagnostic accuracy is lower. Complications include pneumothorax in 20% to 40% of cases and rarely self-limiting hemorrhage and hemoptysis.1,2 There are several guidelines recommending that the procedure be performed by, or under the supervision of, an experienced pulmonologist or radiologist.1–3

Some of the factors influencing the yield and risk of complications of the procedure include the expertise of the operator, site and size of the lesion and distance of the lesion from the chest wall, presence of an on-site pathologist, and underlying pulmonary diseases of the patient.4

Interventional pulmonology is now a well-established subspecialty at most major medical centers in western countries. Primarily, physicians trained in pulmonary disorders perform these procedures. Irrespective of the specialist who performs these procedures, the ability to optimally and safely carry out each of these procedures is ultimately dependent on the training and competence of the operator. The development of initial expertise depends on the residency and fellowship training programs. The American Board of Internal Medicine (ABIM) is responsible for the board certification of the US trained pulmonologists. Currently, the ABIM does not include TTNA/CBN as a required procedure for pulmonary trainees.3 In the United States, only one-third of the pulmonary fellowship programs offer training in TTNA/CBN and only 10% are able to achieve the recommended numbers required for competency.5

At our institution, these procedures are routinely performed either by a trained pulmonologist supervising a fellow or by interventional radiologist (IR). We compared the yield and complications of TTNA/CNB when performed by a pulmonologist versus a dedicated IR.

We hypothesized that the yield for the procedure will be higher and the complications lower when the procedure is performed by a single, trained operator like the IR.

Back to Top | Article Outline


The protocol for this study was approved by the local institutional review board.

We review the medical records of 35 patients who underwent TTNA/CNB at our institution between January 2003 and December 2007. There were no exclusion criteria.

Patients were divided into 2 groups: “pulmonary” when the procedure was performed by a pulmonologist and “radiology” when it was performed by an IR TTNA/CBN.

All procedures are performed under chest computed tomogram (CT) guidance. TTNA/CNBs were performed either by a single IR or a pulmonary fellow supervised by any of 4 pulmonologists with training in TTNA/CNB. At our institution, a pulmonary attending supervises an average of 2 to 4 TTNA/CNBs per year.

Local 1% lidocaine anesthesia was administered in all patients. No sedation was administered. Vital signs with continuous pulse oximetry monitoring were performed during the procedure. A chest radiograph was obtained after the procedure to evaluate for complications.

In both groups, a 19 or 22-gauge Techna-Cut needles (InterV, Medical technology Inc) were used for TTNAs. Core biopsy was performed using an 18 or 19-gauge Quick-Core needle biopsy (Cook Inc; Bloomington).

At our institution, the specimens are reviewed by an on-site pathologist when procedures are performed by a pulmonologist. The pathologist prepares the slides, evaluates quality of the specimen, and renders a preliminary interpretation. Cellblocks from the needle rinses are made in all cases for histopathologic evaluation.

No on-site pathologist was present during the procedures performed by an IR.

Back to Top | Article Outline

Statistical Analysis

Differences between groups were evaluated by t tests for continuous data (age) and by χ2 or Fisher exact test for categorical data. Differences were considered statistically significant at P<0.05.

Back to Top | Article Outline


Study Population

Between 2003 and 2007, 35 patients underwent TTNA and/or CNB; 80% were males. The mean age was 62 years. There were 19 patients in the pulmonary and 16 in the IR group. The characteristics of the 35 patients are presented in Table 1. There was a predominance of smokers in the pulmonary group and this was statistically significant. There was no difference in the radiologic characteristic of the lesions by chest CT (Table 2). In both groups the lesion was large (>3 cm) and abutting the pleura.





Back to Top | Article Outline

TTNA/CBN Results

The main indication for TTNA/CBN in both groups was the evaluation of a lung mass and suspected malignancy. There was no difference in the diagnostic yield or complications for both groups. Of the 35 TTNA and/or biopsy results, 16 were malignant, 7 were suggestive of an inflammatory, benign process, and 12 were nonspecific. The cell types of the 16 malignant specimens were as follows: adenocarcinoma, 6; squamous cell, 5; metastatic carcinoma, 2; small cell, 1: and others, 2. Adenocarcinoma and squamous cell carcinoma were the most common malignancies. The overall yield of the procedure was 66%, with malignancy being the most common diagnosis, seen in 70% of the diagnostic procedures (Table 3).



TTNA was the preferred procedure for pulmonologist compared with radiologists (79% vs. 50%). Twelve patients underwent CNB and TTNA, with 9 being positive; CBN was positive in all 9 patients (Table 4). The number of pleural passes for both groups was 2.4 (range: 1 to 3).



Back to Top | Article Outline

Follow-up Results and Complications

Seven patients had TTNA and/or CNB consistent with inflammatory or benign process; 2 of them underwent further diagnostic procedure confirming the diagnosis of a benign process. The 5 remaining patients had complete resolution of the pulmonary abnormalities after antibiotic treatment.

Ten of the 12 patients with nonconclusive TTNA and/or CBN underwent further diagnostic procedures; a benign etiology was confirmed in 1, and the remaining 9 patients all had malignancy. The most common diagnostic procedure performed was surgical biopsy (Table 5). Of the other 2 patients with a nonconclusive procedure, one died during hospital stay and the other refused further diagnostic procedures.



The most common complication noted was pneumothorax. Four patients (11%) developed a pneumothorax, 2 required chest tube placement.

Back to Top | Article Outline


This study, which compared diagnostic yield and complications of TTNA and/or CBN for pulmonologists versus IRs, demonstrated that there was no difference between these 2 groups.

TTNA has been used in the diagnosis of thoracic lesions for more than 100 years. Over the past 20 years, improvements in technology have led to an explosion in the procedures available for the diagnostic and therapeutic aspects of pulmonary medicine. Interventional pulmonology is an evolving field in pulmonary medicine that focuses on providing consultative and procedural services to patients with pulmonary disorders. The emergence of this discipline has been fueled by the surge in technology, and by the need to maintain procedural competency and expertise in an everexpanding specialty of pulmonary medicine. There are still numerous problems and obstacles to make these services available across the healthcare system. Important issues are expense of the technology and training and maintenance of competences. TTNA and CNB remain an orphan procedure, no longer required by ABIM for pulmonary certification and not consistently incorporated in the procedures under the direction of an interventional pulmonologist. The American College of Chest Physicians (ACCP), the European Respiratory Society, and the American Thoracic Society have published guidelines on interventional procedures with recommendations on the number of TTNA and CNB to be performed to achieve and maintain proficiency. The recommended number is 10 for each technique.1–3,6

The American College of Radiology requires a minimum of 35 image-guided TTNA and CNB to achieve competency.7 Currently, diplomates of the American Board of Radiology are expected to maintain their certification by completing the board's Maintenance of Certification program.8

The British Thoracic Society (BTS) in agreement with the Royal College of Radiologists and the British Society of Interventional Radiology recommends that operators who perform radiologically guided biopsies audit their own practice and monitor their complication rates. These should be similar to, or better than, those from national surveys: pneumothorax (20%), pneumothorax requiring drainage (3%), hemoptysis (5%), and death (<0.1%).9

In a large national survey of practicing pulmonologists, Ernst and collaborators2 reinforced the notion that the overall procedural skills of most pulmonologists are inadequate and or not uniform. Only one-quarter of practicing pulmonologists currently perform all procedures that are required for board certification in pulmonary medicine.2

A survey of 94 pulmonary and critical care fellowship programs in the United States which included university (89%) and community hospitals (11%) revealed that 33% of the programs offered training in TTNA/CNB with only 10% able to achieve the recommended competency number. All fellowships were of 3 years duration and had an average of 14 faculties per program. Forty-nine programs (53%) had a dedicated interventional pulmonologist and 9 had a single pulmonologist supervising more than half of the procedures.5

Our institution is the largest voluntary, not-for-profit healthcare system serving the South and Central Bronx in New York, with 865 acute care beds at 2 major hospital locations. The pulmonary and critical care division has 8 faculty members and 5 pulmonary fellows; 4 of the faculty have training in TTNA/CBN and perform these procedures. The average number of TTNA/CBN per faculty is 1 to 2/y, a number, much lower than recommended to achieve or maintain competence. In a community hospital like ours, it is not feasible to achieve the minimum requirements recommended by the ACCP. But, as recommended by the BTS, the yield and rate of complications is monitored.9 A recent survey of 139 thoracic radiologists practicing in academic (74%) and community based (26%) hospitals in the United States and Canada revealed that approximately 70% perform the procedure under CT guidance. Seventy percent of responders preferred TTNA, whereas 14% preferred CNB and 13% both. An on-site pathologist was available to 73% of responders.10 Another report from the UK again reveals a great variation in practice, with IR performing most of the procedures.11

The choice between TTNA and CNB is controversial and depends upon personal experience and the availability of a pathologist on site. Most of the pulmonologists, including our faculty, are more comfortable with TTNA and have an on-site pathologist.

Our study did not show any difference in the yield of TTNA or CBN when the procedure was performed by an IR without an on-site pathologist or a pulmonologist with an on-site pathologist. The yield for CNB was higher than TTNA alone in both groups, 61% versus 75%, but without reaching statistical significance. Some investigators suggest that when the most probable diagnosis is malignancy and an on-site pathologist is available, TTNA should be the first choice.12,13 However, CBN have been reported to provide higher yield in the specific diagnosis of benign lesions and lymphoma when compared with TTNA, even in the absence of an on-site pathologist.12,14

Some studies suggest that the combined use of TTNA and CNB improves the diagnostic ability of CT-guided lung biopsy, even in small, <10 mm lesions.15,16

The reported accuracy of TTNA and CNB for the diagnosis of pulmonary malignancy or malignant mediastinal lymphadenopathy ranges from 80% to 95%. Negative predictive values range from 84% to 96%, with false-positive rates of 2% to 4%. However, CNB can provide a specific benign diagnosis in 44% to 100% of cases, whereas TTNA has been shown to establish a specific benign diagnosis in only 12% to 47%.1,17,18

The reported diagnostic sensitivity for TTNA/CNB is 68% to 96%, with a specificity of 100%, and an accuracy of 74% to 96% in lesions of all sizes.19,20

In our study, all patients had lesions larger than 3 cm, but the diagnostic accuracy was closer to the accuracy reported for lesions smaller than 3 cm.4,21 One of the reasons could be the fewer procedures performed at our community hospital compared with university hospitals. In a report of 130 TTNAs, the sensitivity for small lesions (<3 cm) was 67% compared with 81% for lesions ≥3 cm.4 Other investigators have found no differences for the diagnostic yield of TTNA between small and larger lesions.19

The overall prevalence of malignancy in those patients with nonspecific TTNA/CNB results and after further surgical biopsy was 90% in our study, a rate similar to the 33% to 100% reported by others.4,22,23

Whereas some investigators found no difference in the complication rate of TTNA compared with CNB, others reported an increased rate of hemorrhage with CNB. The incidence of pneumothorax ranges widely from 6% to 60%, with higher rate if patients have moderate to severe emphysema or when CNB is performed. Hemoptysis is reported in 5% to 15% of cases. Tension pneumothorax and death are rare complications of TTNA.12,14,21,24

A report of 331 CT-guided and 24 fluoroscopically guided lung biopsies performed either by a trained radiologist or radiology residents at different levels of training, revealed that smaller lesion and presence of emphysema strongly correlated with occurrence of pneumothorax. Pneumothorax was reported to be less frequent in lesions abutting the pleura. Lesion location, needle size, number of pleural passes, and level of training did not correlate with rate of pneumothorax.25

In our study, no differences in complication rates were found between the groups.

The limitations of this study include its retrospective nature and absence of an on-site pathologist in the IR group. It is possible that the yield for IR could be increased by the presence of an on-site pathologist. Second, the number of cases was too small to find a significant difference for yield and complication rate. Further, larger prospective, randomized comparative studies with an on-site pathologist in both groups could help to validate our findings.

Back to Top | Article Outline


We could find no other study comparing the yield and complications of TTNA and CNB between a noninterventional pulmonologist and an IR in an inner-city community hospital. We have demonstrated that TTNA with or without CBN remains a useful and safe procedure in the hands of noninterventional pulmonologist and fellows in training. In the setting of community hospitals, neither the pulmonologist nor the IR can perform the required number of procedures to maintain proficiency. The diagnostic yield of the procedure is similar for the IR and the pulmonologist if the latter group uses assistance from an on-site cytopathologist.

The goal of fellowship programs is to ensure that trainees receive the best possible training in both the cognitive disciplines of pathophysiology and procedural skills that define their specialty. Although current consensus guidelines advocate competency numbers that may be too high for community programs, it is clear that a minimum number of procedures or a different competency system should be in place to achieve and maintain competency. Pulmonary training programs could consider limiting the number of faculty performing the procedure or teaming with radiologist to train pulmonary fellows and faculty if feasible. We recommend both TTNA and CNB to maximize the yield of the procedure.

We believe that pulmonary fellowship programs should continue to offer training in TPNA/CNB as long as the faculty has the expertise to do it. Many of our graduates will join small community hospital where there is no interventional radiology or pulmonologist and transferring or referral of patients may not be always possible. Obtaining expertise in CNB is of the outmost importance to maximize the yield of the procedure and decrease the need for further invasive diagnostic procedures. We suggest that if on-site cytologic examination is available pulmonologists can provide as good yield as the IR while performing TTNA and CNB.

Back to Top | Article Outline


1. Bolliger CT, Mathur PN. ERS/ATS statement on interventional pulmonology. Eur Respir J. 2002;19:356–373.
2. Ernst A, Silvestri GA, Johnstone D. ACCP Interventional Chest/Diagnostic Procedures Network Steering Committee Interventional Pulmonary Procedures Guidelines from the American College of Chest Physicians. Chest. 2003;123:1693–1717.
3. Prakash UBS. Guidelines for training and practice of interventional pulmonology by the numbers? J Bronchol. 2003;10.
4. Larscheid RC, Thorpe PE, Scott WJ. Percutaneous transthoracic needle aspiration biopsy. A comprehensive review of its current role in the diagnosis and treatment of lung tumors. Chest. 1998;114:704–709.
5. Pastis JN, Nietert PJ, Silvestri GA, for the American College of Chest Physicians Interventional Chest/Diagnostic Procedures Network Steering Committee Variation in Training for Interventional Pulmonary Procedures Among US Pulmonary/Critical Care Fellowships. A Survey of Fellowship Directors. Chest. 2005;127:1614–1621.
6. Wahidi MM, Herth FJF, Ernst A. State of the art: interventional pulmonology. Chest. 2007;131:261–274.
7. 2006 Practice Guidelines and Technical Standards. American College of Radiology. Available at:
8. Bosma J, Laszakovits D, Hattery R. Self-assessment for maintenance of certification. J Am Coll Radiol. 2007;4:45–52.
9. Manhire A, Charig M, Clelland C, et al. Guidelines for radiologically guided lung biopsy. Thorax. 2003;58:920–936.
10. Avirama G, Schwartzb DS, Meirsdorfa S, et al. Transthoracic needle biopsy of lung masses: a survey of techniques. Clin Radiol. 2005;60:370–374.
11. Richardson CM, Pointon KS, Manhire AR, et al. Percutaneous lung biopsies: a survey of UK practice based on 5444 biopsies. Br J Radiol. 2002;75:731–735.
12. Greif J, Marmur S, Schwarz Y, et al. Percutaneous core cutting needle biopsy compared with fine-needle aspiration in the diagnosis of peripheral lung malignant lesions: results in 156 patients. Cancer. 1998;84:144–147.
13. Kucuk C, Yilmaz A, Akkaya E. Computed tomography-guided transthoracic fine-needle aspiration in diagnosis of lung cancer: a comparison of single-pass needle and multiple-pass coaxial needle systems and the value of immediate cytological assessment. Respirology. 2004;9:392–396.
14. Arakawa H, Nakajima Y, Kurihara Y, et al. CT guided transthoracic needle biopsy: a comparison between automated biopsy gun and fine needle aspiration. Clin Radiol. 1996;51:503–506.
15. Yamagami T, Iida S, Kato T, et al. Combining fine-needle aspiration and core biopsy under CT fluoroscopy guidance: a better way to treat patients with lung nodules? AJR. 2003;180:811–815.
16. Gong Y, Sneige N, Guo M, et al. Transthoracic fine-needle aspiration vs concurrent core needle biopsy in diagnosis of intrathoracic lesions. A retrospective comparison of diagnostic accuracy. Am J Clin Pathol. 2006;125:438–444.
17. Ghaye B, Dondelinger RF. Imaging guided thoracic interventions. Eur Respir J. 2001;17:507–528.
18. 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.
19. Bilaceroglu S, Chhajed P, Mavis A, et al. Computed tomography-guided transthoracic needle aspiration in diagnosing and staging hilar and mediastinal masses of lung cancer after negative bronchoscopy. J Bronchol. 2007;14:149–155.
20. Kanne J. Transthoracic needle biopsy [editorial]. J Bronchol. 2007;14.
21. Li H, Boiselle PM, Shepard JO, et al. Diagnostic accuracy and safety of CT-guided percutaneous needle aspiration biopsy of the lung: comparison of small and large pulmonary nodules. AJR Am J Roentgenol. 1996;167:105–109.
22. Khouri NF, Stitik FP, Erozan YS, et al. Transthoracic needle aspiration biopsy of benign and malignant lung lesions. AJR Am J Roentgenol. 1985;144:281–288.
23. Miller KS, Fish BG, Stanely JH, et al. Predication of pneumothorax rate in percutaneous needle aspiration of the lung. Chest. 1988;93:742–745.
24. Kazerooni EA, Lim FT, Mikhail A, et al. Risk of pneumothorax in CT-guided transthoracic aspiration biopsy of the lung. Radiology. 1996;198:371–375.
25. Cox JE, Chiles C, McManus CH, et al. Transthoracic needle aspiration biopsy: variables that affect risk of pneumothorax. Radiology. 1999;212:165–168.

transthoracic needle aspiration; core needle biopsy; resident training; interventional radiology

© 2008 Lippincott Williams & Wilkins, Inc.