- Email: [email protected]
Correlation of fine-needle aspiration with fluorodeoxyglucose positron emission tomography and ultrasonography imaging in head and neck lymph nodes
Journal of the American Society of Cytopathology (2015) 4, 246e252
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.jascyto.org/
ORIGINAL ARTICLE
Correlation of fine-needle aspiration with fluorodeoxyglucose positron emission tomography and ultrasonography imaging in head and neck lymph nodes Sule Canberk, MDa, Zubair Baloch, MDb, Lisa Jones, MDc, Vidhu Kaushik, MB, BSd, Prabodh K. Gupta, MDb,* a
Department of Pathology and Cytopathology, Haydarpasa Numune Education and Research Hospital, Istanbul, Turkey b Department of Pathology and Laboratory Medicine, UPENN Medical Center, Perelman School of Medicine, 6 Founders Pavilion, 3400 Spruce Street, Philadelphia, Pennsylvania c Department of Radiology, Body Imaging, and MRI, UPENN Medical Center, Perelman School of Medicine, Philadelphia, Pennsylvania d Department of Pathology and Laboratory Medicine, Mayo Clinic Hospital, St. Mary’s Campus, Rochester, Minnesota Received 30 December 2014; received in revised form 23 February 2015; accepted 23 February 2015
KEYWORDS Fine-needle aspiration; Fluorodeoxyglucose positron emission tomography; Ultrasonography imaging; Head and neck; Lymph node
Introduction Fluorodeoxyglucose positron emission tomography (FDG-PET) is a well-established tracer technique, particularly useful for oncologic conditions. However, in the head and neck region, this modality may have limitations due to the complex anatomy and high frequency of normal variants in FDG-PET uptake resulting in diagnostic errors. Hence, FDG-PET results often require additional diagnostic investigations including ultrasonography (US), and guided fine-needle aspirations (FNAs) in order to improve diagnosis and patient care. Materials and methods A total of 87 cases of head and neck lymph node FNA were accessioned in the cytopathology laboratory between 2010 and 2013, 75 cases with corresponding standard uptake values (SUVs) were selected to form the current study cohort. Patient demographics, primary tumor site, size of the biopsied lymph node, PET SUV, and US findings, as well as FNA diagnoses and surgical follow-ups were reviewed. SUVs were grouped into 5 ranges: A (2.2-2.9), B (3.0-4.9), C (5.0-6.9), D (7.0-9.9), and E (10.0 þ).
*Corresponding author: Prabodh K. Gupta, MD; Department of Pathology and Laboratory Medicine, UPENN Medical Center, Perelman School of Medicine, 6 Founders Pavilion, 3400 Spruce Street, Philadelphia, PA 19104; Tel.: þ1 (215) 662-3238. E-mail address: [email protected] (P.K. Gupta). 2213-2945/$36 Ó 2015 American Society of Cytopathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jasc.2015.02.007
FNA with FDG-PET and US imaging in head and neck lymph nodes
247
Results The SUV cutoff of 5.0 was seen in the groups C, D, and E, which increased the probability of malignancy in the FNA specimens from 67% to 100%. The average size of the lymph nodes was 18.57 mm (range 6-41 mm); it was 13.4, 14.2, 19.7, 18.4, and 20.2 for groups A, B, C, D, and E, respectively. The lymph node measurements represented similarities in groups A and B (13.4, 14.2), and groups C, D, and E (19.7, 18.4, 20.2). Albeit a few exceptions, an increase in the lymph node size 14.2 to 19.7 (ie, from groups A and B to groups C, D, and E) correlated with SUV and US abnormalities. Abnormal US findings were pronounced (100%) in group E, and ranged between 67% and 73% of the cases in other groups. Conclusions This study documents the usefulness of SUV range in FDG-PET results and not an absolute “cutoff” number for diagnosing suspicious head and neck lymph nodes due to an overlap in the SUVs between malignant and benign lesions. False positive and negative results may occur in hyperplastic and necrotic lymph nodes. An increase in the lymph node size generally compares well with SUVs and US abnormalities; use of US-guided FNAs in these cases can be helpful to achieve definite diagnosis. Ó 2015 American Society of Cytopathology. Published by Elsevier Inc. All rights reserved.
Introduction Fluorodeoxyglucose positron emission tomography (FDGPET) is an imaging modality that detects metabolic alterations in the tissues by fluorodeoxyglucose F 18 (18F-FDG), a glucose analog. This pharmaceutical substance not only mirrors the glucose metabolism, it is also able to quantify glucose uptake by malignant cells when compared with normal tissues and/or benign conditions. In order to differentiate benign from malignant processes, FDG-PET scans can be evaluated qualitatively (visual analysis) or quantitatively by using a standardized uptake value (SUV) that reflects a degree of 18F-FDG uptakes.1,2 In the past few decades, FDG-PET scan has proven itself as a noninvasive useful imaging modality in the diagnosis, staging, and the monitoring of patients for tumor recurrences. However, in the head and neck regions, this procedure may have limitations due to the complex anatomy and high frequency of variations in 18F-FDG uptake.1 Moreover, metabolic and chemical alterations including inflammatory and infectious conditions are frequent causes for the false-positive FDG-PET results.3 Hence, FDG-PET interpretation especially in the head and neck region alone may be fraught with diagnostic pitfalls and additional correlation with (ultrasonography-guided fine-needle aspiration [USG-FNA]), and tissue studies are often needed. The goal of the study was to assess the diagnostic accuracy of FDG-PET-positive head and neck region lymph nodes results using SUV ranges, size of the lymph node, and US and USG-FNA techniques, as well as to identify the common pitfalls in the interpretation of FDG-PET results.
Materials and methods The cytopathology files of the Hospital of the University of Pennsylvania for four years (2010-2013) were searched for USG-FNA biopsies of head and neck lymph nodes. From a total of 87 identified cases, a cohort of 75 patients with available PET-SUVs was selected and included in this study. The following data points were recorded: patient’s
age, sex, primary tumor site, lymph node size, PET-SUV and US findings; FNA diagnoses and histologic follow-ups were also analyzed. The descriptive statistical analysis was performed by employing Microsoft Excel (version 2011). The maximum SUV on PET images was classified into 5 categories based on the following ranges: A (2.2-2.9), B (3.0-4.9), C (5.0-6.9), D (7.0-9.9), and E (10.0 þ). The US findings were reported as “normal” or “abnormal” based on the assessment of the lymph nodes lesion parameters including size, shape, border, echogenicity, and thickening and symmetry of the capsule. The US findings and SUVs were evaluated in combination with cytopathology and surgical pathology reports. Various lesions were classified as “benign,” “malignant,” and “suspicious” based on the pathology diagnoses.
Results The retrospective computerized search for US-guided FNA biopsies of head and neck region lymph nodes identified 75 patients. These included 52 male and 23 female patients with the average age of 62.3 years (range 18-91 years). Over half, 44 of 75 (59%) of the cases represented metastases from head and neck primaries followed by lung and hematopoietic tumors (Table 1). Corresponding US findings, size of the lymph node, and surgical pathology follow-up (SPFU) were available in 72, 75, and 36 cases, respectively. PET SUVs and USG-FNA findings: The average PETSUV for the 75 cases was 5.86 (range 2.2-14.8): 7.23 (2.614.8) for 39 malignant and 4.3 (2.20-14.50) for 35 benign cases. One 6-mm lymph node with PET SUV of 4.2 was diagnosed as “suspicious for malignancy” by FNA; subsequent biopsy revealed para-cortical hyperplasia. PET SUVs and lymph node size: The average size of 75 lymph nodes was 18.57 mm (range 6-41 mm). It was 18.57 mm (7-41 mm) for 39 malignant, and 14.81 mm (4-31 mm) for 35 benign cases. One specimen reported as suspicious on FNA was 6 mm in size. PET SUVs and SPFU findings: Tissue studies were available in 30 cases. An average PET SUV of 7.7 (range
248 Table 1
S. Canberk et al. Primary tumor sites and patient demographics.
Primary site
Case # (N Z 75)
Sex (M/F)
Mean age (range)
Head and neck Pulmonary Hematopoietic Gastrointestinal Thyroid Cutaneous melanoma Cervix Breast Urinary system Neuroblastoma Unknown primary
44 10 8 3 2 3 1 1 1 1 1
36/8 6/4 4/4 3/0 2/0 1/2 0/1 0/1 1/0 0/1 1/0
62.6 61.5 58.3 61.6 66 54.3 41 61 65 64 62
(46-85) (44-84) (18-91) (54-66) (65-67) (45-66)
Abbreviations: F, female; M, male.
2.60-14.80) for 26 malignant and 6.5 (3.4-14.50) for 4 benign cases was noted. The benign cohort included 1 case with history of Hodgkin disease with SUV of 14.50 that was diagnosed as reactive lymphoid hyperplasia by FNA and biopsy. Excluding this single case from the calculations, the average SUV for benign lymph nodes was 3.8 (range 3.44.2) reflecting a significant difference between malignant and benign cases. Based on the histopathologic follow-up of these 30 cases with accepted SUV cutoff of 5.0; sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) rates were 77% (21 of 27), 90% (9 of 10), 95% (21 of 22), and 60% (9 of 15). An SUV cutoff value of 5.0, as seen in groups C, D, and E, also increased the probability of malignancy in FNA specimens from 67% to 100%. Data about the SUV range in the various groups, FNA diagnoses, US findings, and the average size of the lymph nodes are summarized in Table 2. Based on the FNA diagnoses of malignant and benign (74 cases, 1 suspicious case of 75 cases) with SUVs, the sensitivity, specificity, PPV, and NPV were 77%, 83%, 77%, and 78%, respectively. The cytologic-histologic correlation of 30 specimens with corresponding SUVs (Table 3) had 3 false-negative cases and 1 false-positive case. All false negatives were reported as “lymphocytes present, no carcinoma seen” on FNA and showed metastatic disease; 2 squamous cell
Table 2
carcinomas (SCCs) and 1 poorly differentiated carcinoma (PDC) were reported on SPFU. The 1 false-positive case reported as SCC metastasis had extensive necrosis and no viable tumor seen in SPFU. The lymph nodes sizes and SUVs are depicted in Table 4. The 39 FNA cases diagnosed as malignant on FNA with corresponding SUVs included: 12 PDCs, 11 SCCs, 6 adenocarcinomas (ADCs), 2 thyroid carcinomas, 7 lymphomas, and 1 melanoma. Correlation among the malignant FNA diagnoses, average SUV, and US findings are depicted in Table 5. The highest average SUV was observed in the lymph node metastases from thyroid carcinoma followed by lymphoma, SCC, ADC, and PDC, respectively. Table 6 documents the malignant cases with low (<5.0) SUVs in groups A and B and benign cases with high (>5.0) SUVs in groups C, D, and E. Five of the 6 PDC cases diagnosed on USG-FNA had SUV <5.0. These included cases of lung (2 non-SCCs), and 1 each of nasopharyngeal, cervical carcinoma and olfactory neuroblastoma. The 1 PDC with no past history of malignancy was found to have a colonic adenocarcinoma.
Discussion As a tissue imaging modality, FDG-PET permits correlation of both anatomic and metabolic parameters. It is especially
SUVs versus FNA diagnoses, US findings and the average size of lymph node size.
SUV range
FNA benign
FNA malignant
US abnormal
Size (mm)
A (2.2-2.9) B (3.0-4.9)a C (5.0-6.9) D (7.0-9.9) E (10þ)
78 67 33 20 11
22 30 67 80 89
57 66 67 73 100
13.4 14.2 19.7 18.4 20.2
(7/9) (20/29)a (4/12) (3/15) (1/9)b
(2/9) (9/29)a (8/12) (12/15) (8/9)
Abbreviations: FNA, fine-needle aspiration; SUV, standard uptake value; US, ultrasonography. Values are percentages (n diagnosed/n evaluated). a US changes. b One case.
(4/7) (18/27)a (8/12) (11/15) (9/9)
FNA with FDG-PET and US imaging in head and neck lymph nodes Table 3
249
SPFU, FNA diagnoses, and SUVs for 30 cases.
SUV
FNA diagnosis
SPFU diagnosis
2.6 3.3 3.7 3.7 4.2 4.2 4.6 3.4 4.2 5.2 5.6 6.3 5.6 5.2 5.8 7 8.6 7.7 9.7 8.5 9.9 9.6 14.8 14 12 1.01 11.2 14.5 11 10.3
Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/atypical lymphocytes (T cell) Suspicious/atypical lymphocytes present Benign/lymphocytes present Benign/lymphocytes present, no carcinoma Malignant/SCC Benign/acute inflammation/no tumor Malignant/lymphoma Malignant/SLL Benign/lymphocytes, no carcinoma Malignant/PDC Malignant/SCC Malignant/SCC Malignant/ADC Malignant/atypical lymphoid proliferation Malignant/PDC Malignant/SCC Malignant/ADC Malignant/SCC Malignant/SCC Malignant/SLL Malignant/lymphoma Malignant/SCC Malignant/PDC Malignant/papillary carcinoma, tall cell Benign/lymphocytes Malignant/poorly differentiated carcinoma Malignant/carcinoma, oncocytic features
Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/lymphoma, T-cell type Benign/no lymphoma or carcinoma Benign/reactive lymphadenopathy Malignant/SCC Benign/extensive keratin, no viable tumor Malignant/SCC Malignant/lymphoma, marginal zone Malignant/SLL Malignant/PDC Malignant/PDC Malignant/SCC Malignant/SCC Malignant/ADC Malignant/B-cell neoplasm Malignant/PDC Malignant/SCC Malignant/ADC Malignant/SCC Malignant/SCC Malignant/lymphoma, follicular Malignant/lymphoma, Hodgkin Malignant/SCC Malignant/PDC Malignant/TCA Benign/benign Malignant/PDC Malignant/TCA, oncocytic features
Abbreviations: ADC, Adenocarcinoma; PDC, poorly differentiated carcinoma; SCC, squamous cell carcinoma; SLL, small lymphoblastic lymphoma; SPFU, surgical pathology follow-up; TCA, papillary thyroid carcinoma; other abbreviations as in Table 2.
valuable in the lesions of head and neck region, provides initial tumor staging, detects recurrent/primary tumors and metastases, and monitors the success or failure of a specific therapeutic intervention.1 Lymphadenopathy in the head and neck region may originate from known (metastases) or unknown (eg, carcinoma, lymphoma) malignancy and reactive response to benign processes. Metastatic lymph nodes are generally evaluated to stage malignancy and plan appropriate courses of management.4 Various screening modalities such as US,
Table 4
computerized tomography, and FDG-PET scans have been used for the evaluation of lymph nodes. The sensitivity and specificity of these procedures have been variable. Additionally, FNA is commonly used to assess lymph nodes in the head and neck region that are deemed suspicious by radiologic imaging. In the present study, we report our institutional experience with the utility of USG-FNAs as an adjuvant to FDGPET in the diagnosis of lymphadenopathy in the head and neck region. Our data demonstrates exponentially increased
Details of the false positive and false-negative cases.
SUV
Size (mm)
FNA diagnosis
SPFU diagnosis
4.6 3.4 4.2 6.3
15 15 23 19
Benign/lymphocytes, no carcinoma Malignant/SCC Benign/acute inflammation/no tumor Benign/lymphocytes, no carcinoma
Malignant/SCC Benign/extensive keratin, no tumor Malignant/SCC Malignant/PDC
Abbreviations as in Tables 2 and 3.
250
S. Canberk et al.
Table 5 Correlation between the FNA malignant tumor diagnoses, mean SUV and US findings. FNA diagnoses
Mean SUV
Abnormal US (%)
PDC (n Z 12) SCC (n Z 11) Lymphoma (n Z 7) ADC (n Z 6) TCA (n Z 2) Melanoma (n Z 1)
6.07 7.35 9.19 6.65 10.75 2.6
92 91 100 50 100 100
Abbreviations as in Tables 2 and 3.
malignancy rates beginning with the SUVs >5.0, as seen in groups C, D, and E. It also demonstrates direct correlation with US findings and lymph node size with malignancy. Several studies suggest that the best cutoff for SUV to represent malignancy is 2.5.1,5,6 Collins et al6 evaluated FNA specimens concurrently with FDG-PET scans on 28 patients with known head and neck malignancies; both techniques proved to be highly accurate in the detection of locally recurrent and metastatic neoplasia. This study had no SUVs but only with or without evidence of hypermetabolism in suspicious masses. The investigators reported an overall sensitivity of 94%. Since the initial report by Pansare et al,5 which compared PET SUV with FNA results, there are a limited number of papers in English that discuss this comparison. In this study of 83 cases based on FNAs, these investigators agreed to the SUVs 2.5 or more as a representative threshold for malignancy. The results showed high PPV (87%) but low NPV (56%). These investigators reported the overall diagnostic accuracy of 78%. In the study by Ciocca et al7 based on the correlation of PET with FNA findings in head and neck malignancies, there were only 2 false-positive results when the optimal cutoff value of SUV was set as 6.0. These 2 cases demonstrated reactive lymphoid hyperplasia on FNA. The sensitivity and specificity was reported as 90% and 80%, respectively.7 The size of the involved lymph node was not available. Our 1 false-positive case was 3.9 mm with SUV of 8.6 (Fig. 1). Based on our limited observations,
Table 6
caution needs to be exercised in the interpretation of small (<6 mm) lymph nodes with SUV of <5. A single case of Hodgkin disease had SUV of 14.50. The FNA did not reveal any true Reed-Sternberg cells. More similar cases need to be studied before a meaningful conclusion can be drawn. Necrotic material can generate low SUV and caution should be used in the interpretation of cystic lesions with low SUV (Fig. 2). The findings reported by Rosen et al8 are similar to those by Pansare et al.5 These both studies comprise a heterogeneous group of cases representing mostly lung and lymph nodes. The cutoff value of SUV 2.5 was used to discriminate benign from malignant lesions. Differently from Pansare et al,5 Rosen et al8 employed the qualitative evaluation for the PET findings. The results were assigned to 4 categoriesdpositive, negative, suspicious, and indeterminatedbased on the cutoff value of SUV as 2.5 and concurrent PET impression as a visually recognizable focal area of hypermetabolism. The overall sensitivity and specificity for SUV-correlated FNA results were 84% and 52%, respectively. The overall sensitivity and specificity were 100% and 75% based only on PET findings.8 Determination of a cutoff value for PET studies is quite debatable in the published reports. The sensitivity, specificity, PPV, and NPV show striking heterogeneity, varying according to the method of the study, number of cases, and anatomic sites.7 Interestingly, the study by Nguyen et al9 is a good example for the variations in cutoff values that demonstrated 2 different SUV thresholds in 4 anatomic sites including pulmonary nodules (42 cases), mediastinal lymph nodes (42 cases), cervical lymph nodes (65 cases), and adrenal glands (72 cases). Our study comprises a large number of cases from a specific head and neck region, which demonstrates the value of SUV range instead of the cutoff point. Similar observations were made by Moloney et al10 who documented the superiority of SUV range instead of a determined cutoff value in mediastinal lymph nodes staging of non-small cell lung carcinoma in 92 patients. This present study has assessed the SUV ranges with the FNA results and correlated the observations with US findings and the size of the lymph nodes. The malignancy
Malignant and benign lesions in groups A and B and groups C, D, and E.
FNA diagnosis Groups A and B Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/atypical lymphocytes (T cell) Malignant/SCC Groups C, D, and E Benign/lymphocytes, no carcinoma Benign/lymphocytes, no lymphoma
SPFU diagnosis Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/lymphoma, T-cell type Benign/extensive keratin, NTS Malignant/PDC Benign/benign
Abbreviations: NTS, no tumor seen; other abbreviations as in Tables 2, 3, and 4.
SUV 2.6 3.3 3.7 3.7 3.4 6.3 14.5
FNA with FDG-PET and US imaging in head and neck lymph nodes
251
Figure 1 False-positive positron emission tomography (PET) and ultrasonography (US) of a 58-year-old man with history of right tonsillar squamous cell carcinoma. A, Coronal PET computed tomography (CT) fusion image of the chest reveals a fluorodeoxyglucose (FDG)-avid right axillary lymph node (arrow) (standard uptake value [SUV] 8.6). B, Grayscale US image reveals a mildly enlarged 1.4 2.4 3.9 cm lymph node that is otherwise morphologically normal, with a fatty hilum (see arrow). The lymph node was biopsied twice, each time revealing a “mixed population of lymphocytes, no carcinoma seen.” The SUV of this lymph node decreased on follow up PET-CT scans, and 4 years later, the patient has no imaging evidence of recurrence.
ratios of the 5 groups (A [2.2-2.9], B [3.0-4.9], C [5.0-6.9], D [7.0-9.9], and E [10.0 þ]) of SUVs were 22%, 30%, 67%, 80%, and 89%, respectively. The concurrent US findings were in concordance with PET scan data. The average size of the lymph nodes according to those of above-defined groups were 13.4 mm, 14.2 mm, 19.7 mm, 18.4 mm, and 20.2 mm, respectively. Size of the lymph nodes showed no difference in group A-B versus group CD-E; however, from A-B to C-D-E there was a remarkable difference: 14.2 to 19.7. Interestingly, among all malignant cases; the lowest average SUVs were noted in PDC (12 cases), ADC (6 cases), and SCC (11 cases); 6.07, 6.65, and 7.35, respectively, whereas cases of thyroid carcinoma metastases (2 cases) and lymphoma (7 cases) had the highest average SUV as 10.75 and 9.19, respectively. It appears that among cases with previous history of PDC, a
low SUV may be an important observation for a possible metastasis. There were no similarly distinctive features for other malignancies studied in this report. Some studies have reported that well differentiated and small size of tumor may yield a lower FDG uptake.4,11 Our data support this conclusion with the lowest SUV ranges in ADC and SCC cases and the average size of the lymph nodes 13 mm and 14 mm in groups A and B. However, the lowest average SUV was also found in PDC. We believe, that this may be due to the extensive tumor necrosis commonly encountered in cases of PDC.8 Sampling errors in the FNA specimens can result in a false-negative diagnosis. It has been shown that FDG-PET can be associated with high false-negative and false-positive rates due to the size/volume of the tumor, necrotic lymph nodes, and reactive/inflammatory changes.2,12,13
Figure 2 True positive PETþUS but false-negative fine-needle aspiration (FNA) of a 63-year-old female with adenocarcinoma of the right orbit. A, Coronal fused PET-CT image of the neck demonstrates a cluster of FDG-avid right level II lymph nodes (arrow) (SUV 6-6.2). B, Grayscale US image demonstrates an abnormally enlarged heterogeneous lymph node with loss of the fatty hilum. The lymph node was hypervascular on color Doppler (not shown). FNA revealed “lymphocytes present, no carcinoma seen.” Subsequent right neck dissection revealed poorly differentiated carcinoma. Abbreviations as in Figure 1.
252
Conclusions As a screening technique, FDG-PET permits coregistration of anatomic and metabolic data. This method appears to be invaluable in the head and neck region and provides initial tumor staging; recurrent/primary tumor detection, metastasis, and monitoring effectiveness of a specific therapy regimen. To reduce false-negative and false-positive rates of FDGPET, it is important to have a threshold range for FGD-PET values instead of an absolute cutoff point. This may serve as a better indication for FNA as a diagnostic procedure. One must be aware of the false-positive FDG-PET results due to degenerative and reactive processes following therapy. Therefore, adequate sampling of a suspicious FDG-PET lesion is warranted to arrive at definite diagnosis.
Funding sources No specific funding was disclosed.
Conflict of interest disclosures The authors made no disclosures.
Ethic statement by all authors This study was conducted with approval from the Institutional Review Boards of all the institutions associated with this study as applicable. Authors take responsibility to maintain relevant documentation in this respect.
References 1. Schoder H, Yeung HW. Positron emission imaging of head and neck cancer, including thyroid carcinoma. Semin Nucl Med. 2004;34: 180e197.
S. Canberk et al. 2. Abouzied MM, Crawford ES, Nabi HA. 18F-FDG imaging: pitfalls and artifacts. J Nucl Med Technol. 2005;33:145e155. quiz 162e3. 3. Schoder H, Yeung HW, Gonen M, Kraus D, Larson SM. Head and neck cancer: clinical usefulness and accuracy of PET/CT image fusion. Radiology. 2004;231:65e72. 4. Rusthoven KE, Koshy M, Paulino AC. The role of fluorodeoxyglucose positron emission tomography in cervical lymph node metastases from an unknown primary tumor. Cancer. 2004;101: 2641e2649. 5. Pansare V, Bandyopadhyay S, Feng J, et al. Fine needle aspiration outcomes of masses detected by positron emission tomography: correlation with standard uptake value. Acta Cytol. 2007;51: 509e516. 6. Collins BT, Gardner LJ, Verma AK, Lowe VJ, Dunphy FR, Boyd JH. Correlation of fine needle aspiration biopsy and fluoride-18 fluorodeoxyglucose positron emission tomography in the assessment of locally recurrent and metastatic head and neck neoplasia. Acta Cytol. 1998; 42:1325e1329. 7. Ciocca V, Miller MC, Keane WM, Bibbo M. Correlation of positron emission tomography with fine needle aspiration biopsies in head and neck malignancy. Acta Cytol. 2010;54:5e11. 8. Rosen D, Herrington B, Bhargava P, Laucirica R, Verstovsek G. Correlation of tissue biopsy and fine needle aspiration cytology with positron emission tomography results. Patholog Res Int. 2011;2011: 323051. 9. Nguyen NC, Kaushik A, Wolverson MK, Osman MM. Is there a common SUV threshold in oncological FDG PET/CT, at least for some common indications? A retrospective study. Acta Oncol. 2011;50: 670e677. 10. Moloney F, Ryan D, McCarthy L, et al. Increasing the accuracy of 18F-FDG PET/CT interpretation of “mildly positive” mediastinal nodes in the staging of non-small cell lung cancer. Eur J Radiol. 2014;83:843e847. 11. Kole AC, Nieweg OE, Pruim J, et al. Detection of unknown occult primary tumors using positron emission tomography. Cancer. 1998;82: 1160e1166. 12. Stoeckli SJ, Haerle SK, Strobel K, Haile SR, Hany TF, Schuknecht B. Initial staging of the neck in head and neck squamous cell carcinoma: a comparison of CT, PET/CT, and ultrasound-guided fine-needle aspiration cytology. Head Neck. 2012;34:469e476. 13. Hellwig D, Graeter TP, Ukena D, et al. 18F-FDG PET for mediastinal staging of lung cancer: which SUV threshold makes sense? J Nucl Med. 2007;48:1761e1766.
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.jascyto.org/
ORIGINAL ARTICLE
Correlation of fine-needle aspiration with fluorodeoxyglucose positron emission tomography and ultrasonography imaging in head and neck lymph nodes Sule Canberk, MDa, Zubair Baloch, MDb, Lisa Jones, MDc, Vidhu Kaushik, MB, BSd, Prabodh K. Gupta, MDb,* a
Department of Pathology and Cytopathology, Haydarpasa Numune Education and Research Hospital, Istanbul, Turkey b Department of Pathology and Laboratory Medicine, UPENN Medical Center, Perelman School of Medicine, 6 Founders Pavilion, 3400 Spruce Street, Philadelphia, Pennsylvania c Department of Radiology, Body Imaging, and MRI, UPENN Medical Center, Perelman School of Medicine, Philadelphia, Pennsylvania d Department of Pathology and Laboratory Medicine, Mayo Clinic Hospital, St. Mary’s Campus, Rochester, Minnesota Received 30 December 2014; received in revised form 23 February 2015; accepted 23 February 2015
KEYWORDS Fine-needle aspiration; Fluorodeoxyglucose positron emission tomography; Ultrasonography imaging; Head and neck; Lymph node
Introduction Fluorodeoxyglucose positron emission tomography (FDG-PET) is a well-established tracer technique, particularly useful for oncologic conditions. However, in the head and neck region, this modality may have limitations due to the complex anatomy and high frequency of normal variants in FDG-PET uptake resulting in diagnostic errors. Hence, FDG-PET results often require additional diagnostic investigations including ultrasonography (US), and guided fine-needle aspirations (FNAs) in order to improve diagnosis and patient care. Materials and methods A total of 87 cases of head and neck lymph node FNA were accessioned in the cytopathology laboratory between 2010 and 2013, 75 cases with corresponding standard uptake values (SUVs) were selected to form the current study cohort. Patient demographics, primary tumor site, size of the biopsied lymph node, PET SUV, and US findings, as well as FNA diagnoses and surgical follow-ups were reviewed. SUVs were grouped into 5 ranges: A (2.2-2.9), B (3.0-4.9), C (5.0-6.9), D (7.0-9.9), and E (10.0 þ).
*Corresponding author: Prabodh K. Gupta, MD; Department of Pathology and Laboratory Medicine, UPENN Medical Center, Perelman School of Medicine, 6 Founders Pavilion, 3400 Spruce Street, Philadelphia, PA 19104; Tel.: þ1 (215) 662-3238. E-mail address: [email protected] (P.K. Gupta). 2213-2945/$36 Ó 2015 American Society of Cytopathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jasc.2015.02.007
FNA with FDG-PET and US imaging in head and neck lymph nodes
247
Results The SUV cutoff of 5.0 was seen in the groups C, D, and E, which increased the probability of malignancy in the FNA specimens from 67% to 100%. The average size of the lymph nodes was 18.57 mm (range 6-41 mm); it was 13.4, 14.2, 19.7, 18.4, and 20.2 for groups A, B, C, D, and E, respectively. The lymph node measurements represented similarities in groups A and B (13.4, 14.2), and groups C, D, and E (19.7, 18.4, 20.2). Albeit a few exceptions, an increase in the lymph node size 14.2 to 19.7 (ie, from groups A and B to groups C, D, and E) correlated with SUV and US abnormalities. Abnormal US findings were pronounced (100%) in group E, and ranged between 67% and 73% of the cases in other groups. Conclusions This study documents the usefulness of SUV range in FDG-PET results and not an absolute “cutoff” number for diagnosing suspicious head and neck lymph nodes due to an overlap in the SUVs between malignant and benign lesions. False positive and negative results may occur in hyperplastic and necrotic lymph nodes. An increase in the lymph node size generally compares well with SUVs and US abnormalities; use of US-guided FNAs in these cases can be helpful to achieve definite diagnosis. Ó 2015 American Society of Cytopathology. Published by Elsevier Inc. All rights reserved.
Introduction Fluorodeoxyglucose positron emission tomography (FDGPET) is an imaging modality that detects metabolic alterations in the tissues by fluorodeoxyglucose F 18 (18F-FDG), a glucose analog. This pharmaceutical substance not only mirrors the glucose metabolism, it is also able to quantify glucose uptake by malignant cells when compared with normal tissues and/or benign conditions. In order to differentiate benign from malignant processes, FDG-PET scans can be evaluated qualitatively (visual analysis) or quantitatively by using a standardized uptake value (SUV) that reflects a degree of 18F-FDG uptakes.1,2 In the past few decades, FDG-PET scan has proven itself as a noninvasive useful imaging modality in the diagnosis, staging, and the monitoring of patients for tumor recurrences. However, in the head and neck regions, this procedure may have limitations due to the complex anatomy and high frequency of variations in 18F-FDG uptake.1 Moreover, metabolic and chemical alterations including inflammatory and infectious conditions are frequent causes for the false-positive FDG-PET results.3 Hence, FDG-PET interpretation especially in the head and neck region alone may be fraught with diagnostic pitfalls and additional correlation with (ultrasonography-guided fine-needle aspiration [USG-FNA]), and tissue studies are often needed. The goal of the study was to assess the diagnostic accuracy of FDG-PET-positive head and neck region lymph nodes results using SUV ranges, size of the lymph node, and US and USG-FNA techniques, as well as to identify the common pitfalls in the interpretation of FDG-PET results.
Materials and methods The cytopathology files of the Hospital of the University of Pennsylvania for four years (2010-2013) were searched for USG-FNA biopsies of head and neck lymph nodes. From a total of 87 identified cases, a cohort of 75 patients with available PET-SUVs was selected and included in this study. The following data points were recorded: patient’s
age, sex, primary tumor site, lymph node size, PET-SUV and US findings; FNA diagnoses and histologic follow-ups were also analyzed. The descriptive statistical analysis was performed by employing Microsoft Excel (version 2011). The maximum SUV on PET images was classified into 5 categories based on the following ranges: A (2.2-2.9), B (3.0-4.9), C (5.0-6.9), D (7.0-9.9), and E (10.0 þ). The US findings were reported as “normal” or “abnormal” based on the assessment of the lymph nodes lesion parameters including size, shape, border, echogenicity, and thickening and symmetry of the capsule. The US findings and SUVs were evaluated in combination with cytopathology and surgical pathology reports. Various lesions were classified as “benign,” “malignant,” and “suspicious” based on the pathology diagnoses.
Results The retrospective computerized search for US-guided FNA biopsies of head and neck region lymph nodes identified 75 patients. These included 52 male and 23 female patients with the average age of 62.3 years (range 18-91 years). Over half, 44 of 75 (59%) of the cases represented metastases from head and neck primaries followed by lung and hematopoietic tumors (Table 1). Corresponding US findings, size of the lymph node, and surgical pathology follow-up (SPFU) were available in 72, 75, and 36 cases, respectively. PET SUVs and USG-FNA findings: The average PETSUV for the 75 cases was 5.86 (range 2.2-14.8): 7.23 (2.614.8) for 39 malignant and 4.3 (2.20-14.50) for 35 benign cases. One 6-mm lymph node with PET SUV of 4.2 was diagnosed as “suspicious for malignancy” by FNA; subsequent biopsy revealed para-cortical hyperplasia. PET SUVs and lymph node size: The average size of 75 lymph nodes was 18.57 mm (range 6-41 mm). It was 18.57 mm (7-41 mm) for 39 malignant, and 14.81 mm (4-31 mm) for 35 benign cases. One specimen reported as suspicious on FNA was 6 mm in size. PET SUVs and SPFU findings: Tissue studies were available in 30 cases. An average PET SUV of 7.7 (range
248 Table 1
S. Canberk et al. Primary tumor sites and patient demographics.
Primary site
Case # (N Z 75)
Sex (M/F)
Mean age (range)
Head and neck Pulmonary Hematopoietic Gastrointestinal Thyroid Cutaneous melanoma Cervix Breast Urinary system Neuroblastoma Unknown primary
44 10 8 3 2 3 1 1 1 1 1
36/8 6/4 4/4 3/0 2/0 1/2 0/1 0/1 1/0 0/1 1/0
62.6 61.5 58.3 61.6 66 54.3 41 61 65 64 62
(46-85) (44-84) (18-91) (54-66) (65-67) (45-66)
Abbreviations: F, female; M, male.
2.60-14.80) for 26 malignant and 6.5 (3.4-14.50) for 4 benign cases was noted. The benign cohort included 1 case with history of Hodgkin disease with SUV of 14.50 that was diagnosed as reactive lymphoid hyperplasia by FNA and biopsy. Excluding this single case from the calculations, the average SUV for benign lymph nodes was 3.8 (range 3.44.2) reflecting a significant difference between malignant and benign cases. Based on the histopathologic follow-up of these 30 cases with accepted SUV cutoff of 5.0; sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) rates were 77% (21 of 27), 90% (9 of 10), 95% (21 of 22), and 60% (9 of 15). An SUV cutoff value of 5.0, as seen in groups C, D, and E, also increased the probability of malignancy in FNA specimens from 67% to 100%. Data about the SUV range in the various groups, FNA diagnoses, US findings, and the average size of the lymph nodes are summarized in Table 2. Based on the FNA diagnoses of malignant and benign (74 cases, 1 suspicious case of 75 cases) with SUVs, the sensitivity, specificity, PPV, and NPV were 77%, 83%, 77%, and 78%, respectively. The cytologic-histologic correlation of 30 specimens with corresponding SUVs (Table 3) had 3 false-negative cases and 1 false-positive case. All false negatives were reported as “lymphocytes present, no carcinoma seen” on FNA and showed metastatic disease; 2 squamous cell
Table 2
carcinomas (SCCs) and 1 poorly differentiated carcinoma (PDC) were reported on SPFU. The 1 false-positive case reported as SCC metastasis had extensive necrosis and no viable tumor seen in SPFU. The lymph nodes sizes and SUVs are depicted in Table 4. The 39 FNA cases diagnosed as malignant on FNA with corresponding SUVs included: 12 PDCs, 11 SCCs, 6 adenocarcinomas (ADCs), 2 thyroid carcinomas, 7 lymphomas, and 1 melanoma. Correlation among the malignant FNA diagnoses, average SUV, and US findings are depicted in Table 5. The highest average SUV was observed in the lymph node metastases from thyroid carcinoma followed by lymphoma, SCC, ADC, and PDC, respectively. Table 6 documents the malignant cases with low (<5.0) SUVs in groups A and B and benign cases with high (>5.0) SUVs in groups C, D, and E. Five of the 6 PDC cases diagnosed on USG-FNA had SUV <5.0. These included cases of lung (2 non-SCCs), and 1 each of nasopharyngeal, cervical carcinoma and olfactory neuroblastoma. The 1 PDC with no past history of malignancy was found to have a colonic adenocarcinoma.
Discussion As a tissue imaging modality, FDG-PET permits correlation of both anatomic and metabolic parameters. It is especially
SUVs versus FNA diagnoses, US findings and the average size of lymph node size.
SUV range
FNA benign
FNA malignant
US abnormal
Size (mm)
A (2.2-2.9) B (3.0-4.9)a C (5.0-6.9) D (7.0-9.9) E (10þ)
78 67 33 20 11
22 30 67 80 89
57 66 67 73 100
13.4 14.2 19.7 18.4 20.2
(7/9) (20/29)a (4/12) (3/15) (1/9)b
(2/9) (9/29)a (8/12) (12/15) (8/9)
Abbreviations: FNA, fine-needle aspiration; SUV, standard uptake value; US, ultrasonography. Values are percentages (n diagnosed/n evaluated). a US changes. b One case.
(4/7) (18/27)a (8/12) (11/15) (9/9)
FNA with FDG-PET and US imaging in head and neck lymph nodes Table 3
249
SPFU, FNA diagnoses, and SUVs for 30 cases.
SUV
FNA diagnosis
SPFU diagnosis
2.6 3.3 3.7 3.7 4.2 4.2 4.6 3.4 4.2 5.2 5.6 6.3 5.6 5.2 5.8 7 8.6 7.7 9.7 8.5 9.9 9.6 14.8 14 12 1.01 11.2 14.5 11 10.3
Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/atypical lymphocytes (T cell) Suspicious/atypical lymphocytes present Benign/lymphocytes present Benign/lymphocytes present, no carcinoma Malignant/SCC Benign/acute inflammation/no tumor Malignant/lymphoma Malignant/SLL Benign/lymphocytes, no carcinoma Malignant/PDC Malignant/SCC Malignant/SCC Malignant/ADC Malignant/atypical lymphoid proliferation Malignant/PDC Malignant/SCC Malignant/ADC Malignant/SCC Malignant/SCC Malignant/SLL Malignant/lymphoma Malignant/SCC Malignant/PDC Malignant/papillary carcinoma, tall cell Benign/lymphocytes Malignant/poorly differentiated carcinoma Malignant/carcinoma, oncocytic features
Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/lymphoma, T-cell type Benign/no lymphoma or carcinoma Benign/reactive lymphadenopathy Malignant/SCC Benign/extensive keratin, no viable tumor Malignant/SCC Malignant/lymphoma, marginal zone Malignant/SLL Malignant/PDC Malignant/PDC Malignant/SCC Malignant/SCC Malignant/ADC Malignant/B-cell neoplasm Malignant/PDC Malignant/SCC Malignant/ADC Malignant/SCC Malignant/SCC Malignant/lymphoma, follicular Malignant/lymphoma, Hodgkin Malignant/SCC Malignant/PDC Malignant/TCA Benign/benign Malignant/PDC Malignant/TCA, oncocytic features
Abbreviations: ADC, Adenocarcinoma; PDC, poorly differentiated carcinoma; SCC, squamous cell carcinoma; SLL, small lymphoblastic lymphoma; SPFU, surgical pathology follow-up; TCA, papillary thyroid carcinoma; other abbreviations as in Table 2.
valuable in the lesions of head and neck region, provides initial tumor staging, detects recurrent/primary tumors and metastases, and monitors the success or failure of a specific therapeutic intervention.1 Lymphadenopathy in the head and neck region may originate from known (metastases) or unknown (eg, carcinoma, lymphoma) malignancy and reactive response to benign processes. Metastatic lymph nodes are generally evaluated to stage malignancy and plan appropriate courses of management.4 Various screening modalities such as US,
Table 4
computerized tomography, and FDG-PET scans have been used for the evaluation of lymph nodes. The sensitivity and specificity of these procedures have been variable. Additionally, FNA is commonly used to assess lymph nodes in the head and neck region that are deemed suspicious by radiologic imaging. In the present study, we report our institutional experience with the utility of USG-FNAs as an adjuvant to FDGPET in the diagnosis of lymphadenopathy in the head and neck region. Our data demonstrates exponentially increased
Details of the false positive and false-negative cases.
SUV
Size (mm)
FNA diagnosis
SPFU diagnosis
4.6 3.4 4.2 6.3
15 15 23 19
Benign/lymphocytes, no carcinoma Malignant/SCC Benign/acute inflammation/no tumor Benign/lymphocytes, no carcinoma
Malignant/SCC Benign/extensive keratin, no tumor Malignant/SCC Malignant/PDC
Abbreviations as in Tables 2 and 3.
250
S. Canberk et al.
Table 5 Correlation between the FNA malignant tumor diagnoses, mean SUV and US findings. FNA diagnoses
Mean SUV
Abnormal US (%)
PDC (n Z 12) SCC (n Z 11) Lymphoma (n Z 7) ADC (n Z 6) TCA (n Z 2) Melanoma (n Z 1)
6.07 7.35 9.19 6.65 10.75 2.6
92 91 100 50 100 100
Abbreviations as in Tables 2 and 3.
malignancy rates beginning with the SUVs >5.0, as seen in groups C, D, and E. It also demonstrates direct correlation with US findings and lymph node size with malignancy. Several studies suggest that the best cutoff for SUV to represent malignancy is 2.5.1,5,6 Collins et al6 evaluated FNA specimens concurrently with FDG-PET scans on 28 patients with known head and neck malignancies; both techniques proved to be highly accurate in the detection of locally recurrent and metastatic neoplasia. This study had no SUVs but only with or without evidence of hypermetabolism in suspicious masses. The investigators reported an overall sensitivity of 94%. Since the initial report by Pansare et al,5 which compared PET SUV with FNA results, there are a limited number of papers in English that discuss this comparison. In this study of 83 cases based on FNAs, these investigators agreed to the SUVs 2.5 or more as a representative threshold for malignancy. The results showed high PPV (87%) but low NPV (56%). These investigators reported the overall diagnostic accuracy of 78%. In the study by Ciocca et al7 based on the correlation of PET with FNA findings in head and neck malignancies, there were only 2 false-positive results when the optimal cutoff value of SUV was set as 6.0. These 2 cases demonstrated reactive lymphoid hyperplasia on FNA. The sensitivity and specificity was reported as 90% and 80%, respectively.7 The size of the involved lymph node was not available. Our 1 false-positive case was 3.9 mm with SUV of 8.6 (Fig. 1). Based on our limited observations,
Table 6
caution needs to be exercised in the interpretation of small (<6 mm) lymph nodes with SUV of <5. A single case of Hodgkin disease had SUV of 14.50. The FNA did not reveal any true Reed-Sternberg cells. More similar cases need to be studied before a meaningful conclusion can be drawn. Necrotic material can generate low SUV and caution should be used in the interpretation of cystic lesions with low SUV (Fig. 2). The findings reported by Rosen et al8 are similar to those by Pansare et al.5 These both studies comprise a heterogeneous group of cases representing mostly lung and lymph nodes. The cutoff value of SUV 2.5 was used to discriminate benign from malignant lesions. Differently from Pansare et al,5 Rosen et al8 employed the qualitative evaluation for the PET findings. The results were assigned to 4 categoriesdpositive, negative, suspicious, and indeterminatedbased on the cutoff value of SUV as 2.5 and concurrent PET impression as a visually recognizable focal area of hypermetabolism. The overall sensitivity and specificity for SUV-correlated FNA results were 84% and 52%, respectively. The overall sensitivity and specificity were 100% and 75% based only on PET findings.8 Determination of a cutoff value for PET studies is quite debatable in the published reports. The sensitivity, specificity, PPV, and NPV show striking heterogeneity, varying according to the method of the study, number of cases, and anatomic sites.7 Interestingly, the study by Nguyen et al9 is a good example for the variations in cutoff values that demonstrated 2 different SUV thresholds in 4 anatomic sites including pulmonary nodules (42 cases), mediastinal lymph nodes (42 cases), cervical lymph nodes (65 cases), and adrenal glands (72 cases). Our study comprises a large number of cases from a specific head and neck region, which demonstrates the value of SUV range instead of the cutoff point. Similar observations were made by Moloney et al10 who documented the superiority of SUV range instead of a determined cutoff value in mediastinal lymph nodes staging of non-small cell lung carcinoma in 92 patients. This present study has assessed the SUV ranges with the FNA results and correlated the observations with US findings and the size of the lymph nodes. The malignancy
Malignant and benign lesions in groups A and B and groups C, D, and E.
FNA diagnosis Groups A and B Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/atypical lymphocytes (T cell) Malignant/SCC Groups C, D, and E Benign/lymphocytes, no carcinoma Benign/lymphocytes, no lymphoma
SPFU diagnosis Malignant/melanoma Malignant/PDC Malignant/PDC Malignant/lymphoma, T-cell type Benign/extensive keratin, NTS Malignant/PDC Benign/benign
Abbreviations: NTS, no tumor seen; other abbreviations as in Tables 2, 3, and 4.
SUV 2.6 3.3 3.7 3.7 3.4 6.3 14.5
FNA with FDG-PET and US imaging in head and neck lymph nodes
251
Figure 1 False-positive positron emission tomography (PET) and ultrasonography (US) of a 58-year-old man with history of right tonsillar squamous cell carcinoma. A, Coronal PET computed tomography (CT) fusion image of the chest reveals a fluorodeoxyglucose (FDG)-avid right axillary lymph node (arrow) (standard uptake value [SUV] 8.6). B, Grayscale US image reveals a mildly enlarged 1.4 2.4 3.9 cm lymph node that is otherwise morphologically normal, with a fatty hilum (see arrow). The lymph node was biopsied twice, each time revealing a “mixed population of lymphocytes, no carcinoma seen.” The SUV of this lymph node decreased on follow up PET-CT scans, and 4 years later, the patient has no imaging evidence of recurrence.
ratios of the 5 groups (A [2.2-2.9], B [3.0-4.9], C [5.0-6.9], D [7.0-9.9], and E [10.0 þ]) of SUVs were 22%, 30%, 67%, 80%, and 89%, respectively. The concurrent US findings were in concordance with PET scan data. The average size of the lymph nodes according to those of above-defined groups were 13.4 mm, 14.2 mm, 19.7 mm, 18.4 mm, and 20.2 mm, respectively. Size of the lymph nodes showed no difference in group A-B versus group CD-E; however, from A-B to C-D-E there was a remarkable difference: 14.2 to 19.7. Interestingly, among all malignant cases; the lowest average SUVs were noted in PDC (12 cases), ADC (6 cases), and SCC (11 cases); 6.07, 6.65, and 7.35, respectively, whereas cases of thyroid carcinoma metastases (2 cases) and lymphoma (7 cases) had the highest average SUV as 10.75 and 9.19, respectively. It appears that among cases with previous history of PDC, a
low SUV may be an important observation for a possible metastasis. There were no similarly distinctive features for other malignancies studied in this report. Some studies have reported that well differentiated and small size of tumor may yield a lower FDG uptake.4,11 Our data support this conclusion with the lowest SUV ranges in ADC and SCC cases and the average size of the lymph nodes 13 mm and 14 mm in groups A and B. However, the lowest average SUV was also found in PDC. We believe, that this may be due to the extensive tumor necrosis commonly encountered in cases of PDC.8 Sampling errors in the FNA specimens can result in a false-negative diagnosis. It has been shown that FDG-PET can be associated with high false-negative and false-positive rates due to the size/volume of the tumor, necrotic lymph nodes, and reactive/inflammatory changes.2,12,13
Figure 2 True positive PETþUS but false-negative fine-needle aspiration (FNA) of a 63-year-old female with adenocarcinoma of the right orbit. A, Coronal fused PET-CT image of the neck demonstrates a cluster of FDG-avid right level II lymph nodes (arrow) (SUV 6-6.2). B, Grayscale US image demonstrates an abnormally enlarged heterogeneous lymph node with loss of the fatty hilum. The lymph node was hypervascular on color Doppler (not shown). FNA revealed “lymphocytes present, no carcinoma seen.” Subsequent right neck dissection revealed poorly differentiated carcinoma. Abbreviations as in Figure 1.
252
Conclusions As a screening technique, FDG-PET permits coregistration of anatomic and metabolic data. This method appears to be invaluable in the head and neck region and provides initial tumor staging; recurrent/primary tumor detection, metastasis, and monitoring effectiveness of a specific therapy regimen. To reduce false-negative and false-positive rates of FDGPET, it is important to have a threshold range for FGD-PET values instead of an absolute cutoff point. This may serve as a better indication for FNA as a diagnostic procedure. One must be aware of the false-positive FDG-PET results due to degenerative and reactive processes following therapy. Therefore, adequate sampling of a suspicious FDG-PET lesion is warranted to arrive at definite diagnosis.
Funding sources No specific funding was disclosed.
Conflict of interest disclosures The authors made no disclosures.
Ethic statement by all authors This study was conducted with approval from the Institutional Review Boards of all the institutions associated with this study as applicable. Authors take responsibility to maintain relevant documentation in this respect.
References 1. Schoder H, Yeung HW. Positron emission imaging of head and neck cancer, including thyroid carcinoma. Semin Nucl Med. 2004;34: 180e197.
S. Canberk et al. 2. Abouzied MM, Crawford ES, Nabi HA. 18F-FDG imaging: pitfalls and artifacts. J Nucl Med Technol. 2005;33:145e155. quiz 162e3. 3. Schoder H, Yeung HW, Gonen M, Kraus D, Larson SM. Head and neck cancer: clinical usefulness and accuracy of PET/CT image fusion. Radiology. 2004;231:65e72. 4. Rusthoven KE, Koshy M, Paulino AC. The role of fluorodeoxyglucose positron emission tomography in cervical lymph node metastases from an unknown primary tumor. Cancer. 2004;101: 2641e2649. 5. Pansare V, Bandyopadhyay S, Feng J, et al. Fine needle aspiration outcomes of masses detected by positron emission tomography: correlation with standard uptake value. Acta Cytol. 2007;51: 509e516. 6. Collins BT, Gardner LJ, Verma AK, Lowe VJ, Dunphy FR, Boyd JH. Correlation of fine needle aspiration biopsy and fluoride-18 fluorodeoxyglucose positron emission tomography in the assessment of locally recurrent and metastatic head and neck neoplasia. Acta Cytol. 1998; 42:1325e1329. 7. Ciocca V, Miller MC, Keane WM, Bibbo M. Correlation of positron emission tomography with fine needle aspiration biopsies in head and neck malignancy. Acta Cytol. 2010;54:5e11. 8. Rosen D, Herrington B, Bhargava P, Laucirica R, Verstovsek G. Correlation of tissue biopsy and fine needle aspiration cytology with positron emission tomography results. Patholog Res Int. 2011;2011: 323051. 9. Nguyen NC, Kaushik A, Wolverson MK, Osman MM. Is there a common SUV threshold in oncological FDG PET/CT, at least for some common indications? A retrospective study. Acta Oncol. 2011;50: 670e677. 10. Moloney F, Ryan D, McCarthy L, et al. Increasing the accuracy of 18F-FDG PET/CT interpretation of “mildly positive” mediastinal nodes in the staging of non-small cell lung cancer. Eur J Radiol. 2014;83:843e847. 11. Kole AC, Nieweg OE, Pruim J, et al. Detection of unknown occult primary tumors using positron emission tomography. Cancer. 1998;82: 1160e1166. 12. Stoeckli SJ, Haerle SK, Strobel K, Haile SR, Hany TF, Schuknecht B. Initial staging of the neck in head and neck squamous cell carcinoma: a comparison of CT, PET/CT, and ultrasound-guided fine-needle aspiration cytology. Head Neck. 2012;34:469e476. 13. Hellwig D, Graeter TP, Ukena D, et al. 18F-FDG PET for mediastinal staging of lung cancer: which SUV threshold makes sense? J Nucl Med. 2007;48:1761e1766.