Efficacy and pitfalls of fine needle aspiration in the diagnosis of neck masses

P. Sheahan*, J. Fitzgibbon, G. O’Leary*, G. Lee
*Department of Otolaryngology, Head and Neck Surgery, South Infirmary, Victoria Hospital, Cork, Ireland Mercy Hospital, Cork, Ireland

Correspondence to: P. Sheahan, 17 Hampton Square, Navan Road, Dublin 7, Ireland Email: sheahanp@eircom.net

Introduction Materials and Methods Results

Discussion References

Keywords: Fine needle aspiration, carcinoma, cyst, neck
Surg J R Coll Surg Edinb Irel., 2 June 2004, 152-156

Background: Fine needle aspiration biopsy (FNAB) is now well established in the assessment of cervical masses. The purpose of the present study is to review the efficacy of this procedure, as well as to identify any pitfalls that may limit its usefulness. Methods: One hundred and ninety aspirations of neck masses performed over a recent five-year period were reviewed. The definitive diagnosis of the mass was determined in each case by review of the patients’ case notes. Results: Thirty seven per cent of all neck lumps were malignant. The most common cause for a false-negative result, in the case of a carcinomatous mass when an adequate sample had been obtained, was a cystic neoplasm. One quarter of all cystic lateral cervical masses not considered suspicious for malignancy by FNAB turned out to be malignant. Conclusion: Repeating FNAB in cases where the original result is negative for carcinoma may increase the sensitivity of FNAB in the detection of cystic carcinomas

INTRODUCTION
The differential diagnosis of a neck lump covers a broad spectrum of diseases with differing implications for management. Common causes of a neck lump include reactive lymphadenopathy, inflammatory salivary gland enlargement, thyroid gland masses, branchial cysts and benign neoplasms.¹ However, over one half of asymmetrical neck masses in adults are reported to be malignant, either primary (usually lymphoma), or secondary (usually carcinoma).² Therefore, the accurate diagnosis of the nature of a neck swelling is of paramount importance.

The gold-standard procedure for the diagnosis of a neck mass is open biopsy of the mass with histological examination of the excised tissue.³ However, open biopsy of a metastatic cervical mass prior to definitive treatment of the neck (usually by radical neck dissection) in patients with metastatic cervical carcinoma has been reported to lead to a higher incidence of wound complications, regional neck recurrence and distant metastases, than in patients who have no biopsy performed prior to definitive treatment.⁴ Likewise, open biopsy of parotid gland neoplasms prior to definitive treatment (usually by superficial parotidectomy) has been reported to lead to an increased risk of tumour implantation of the wound and, consequently, an increased rate of local recurrence.³ For this reason, open biopsy of a neck mass of unknown aetiology should not be undertaken until the possibility of carcinoma or salivary gland tumour has been excluded.

Fine needle aspiration biopsy (FNAB) is the study of cells aspirated from a mass using a small gauge needle.⁵ It is an easy procedure to perform, can detect the presence of carcinoma, salivary gland tumours and other neoplasms, with a high degree of accuracy.^3,6 Its main advantages are that it is minimally invasive, has virtually no contradictions and does not violate the neck, so allowing the early diagnosis of a neck mass without risking seeding of malignant cells.^6,7 Furthermore, complications of FNAB are rare and the procedure causes minimal distortion of lymph node architecture, and so interferes minimally with subsequent histological examination of excised lymph nodes.³

Fine needle aspiration biopsy has had a considerable impact on the management of patients presenting with neck lumps. However, the unexpected finding of malignancy in an excised neck mass, which pre-operatively had been considered benign, although uncommon, is still encountered on occasion. Such a failure to identify malignancy pre-operatively may have some important consequences. In the first instance, urgent cases may be dealt with non-urgently, leading to a delay in diagnosis and, thus, to a delay in definitive treatment. In addition, the definitive treatment of such patients may entail additional morbidity. Thus, patients with unsuspected squamous cell carcinoma may well require radiotherapy, in addition to surgery as part of their definitive treatment, on account of their necks having been violated prior to this definitive treatment. With these considerations in mind, the present study was undertaken in order to try to identify the reasons for failing to make a correct diagnosis of malignancy using FNAB, as well as identify any means of avoiding these pitfalls.

MATERIALS AND METHODS
The records of 147 adult patients, undergoing a total of 190 FNABs of neck masses, on 153 separate occasions, over a five-year period between 1995 and 2000, were reviewed. The indication for performing FNAB, and the results of this procedure were recorded. The definitive diagnosis of the neck lump in each case was determined by examination of the case notes of all the patients. In cases where the neck lump was subsequently removed, the diagnosis of the lump was established by histological examination of the excised tissue. In cases where the neck lump was not removed, FNAB had suggested a reactive or inflammatory process and where the swelling subsequently resolved without recurrence, then the cause of the swelling was determined to be inflammatory. Finally, in cases where the neck lump was not removed, a diagnosis of carcinoma was suggested by FNAB and where the results of imaging studies and the subsequent clinical course of the patient were consistent with this diagnosis, then the cause of the lump was determined to be carcinoma.

The accuracy of FNAB in the diagnosis of the various pathological diagnoses was determined. In cases where FNAB failed to correctly diagnose the nature of the lump, then the reasons for this failure were examined.

RESULTS

Accuracy of FNAB
Fine needle aspiration biopsy was diagnostic of malignancy in 37 of the 42 cases of squamous cell carcinoma. In five cases, repeat aspiration was necessary after initial aspiration yielded an inadequate sample or was inconclusive. In three of these, initial FNAB had been considered inconclusive, and in the other two, the initial aspirate had been considered inadequate. Four of these five cases undergoing repeat aspiration were cystic metastases (Figure 1). Fine needle aspiration biopsy was specifically diagnostic of squamous carcinoma in 33 cases; the exact type of malignant cell could not be determined in the other four (three metastatic from poorly differentiated nasopharyngeal squamous carcinoma, one metastatic from parotid squamous carcinoma) because the cells were too poorly differentiated.

Fine needle aspiration biopsy failed to diagnose malignancy in five of the 42 cases of squamous carcinoma. In three of these the aspirate obtained by FNAB was considered inadequate, precluding diagnostic cytology. The other two cases were falsely negative for malignancy. Both of these occurred with cystic metastases.

Excluding the three cases where inadequate samples had been obtained, the sensitivity of FNAB in detecting malignancy with squamous cell carcinoma was 95% (37/39) (Table 2). When cystic metastases are excluded, the sensitivity of FNAB in the detection of carcinoma in solid neck lumps was 100%.

Other epithelial malignancies
Fine needle aspiration biopsy was diagnostic of malignancy in all three cases of adenocarcinoma. Repeat aspiration was necessary in one patient, after initial biopsy was considered inadequate. It was also diagnostic of malignancy in the single cases of malignant melanoma, mucoepidermoid carcinoma, papillary carcinoma, and follicular carcinoma. However, FNAB, failed to diagnose malignancy in the single cases of large cell carcinoma, adenoid cystic carcinoma, and acinic cell carcinoma. Both the acinic cell carcinoma and the adenoid cystic carcinoma were cystic neoplasms.

Figure 1: FNAB from a cystic squamous cell carcinoma

The overall sensitivity of FNAB in the diagnosis of malignant epithelial neoplasms was 90% (Table 2). Of note, was the finding that four of the five epithelial malignancies that FNAB had failed to correctly diagnose, despite adequate sampling, were cystic.

Lymphoma
The diagnosis of lymphoma was confirmed histologically in all seven cases. In five (71%) of these, FNAB was suspicious (but not diagnostic) of lymphoma. In the other two cases (29%), FNAB was not considered to be suspicious for lymphoma (but not excluding lymphoma as a diagnosis) (Table 2).

Pleomorphic salivary adenoma
Fine needle aspiration biopsy was diagnostic of pleomorphic salivary adenoma in 11 of the 13 cases. In two cases, repeat aspiration was necessary to arrive at the diagnosis, after initial FNAB had been reported as being inadequate. In the other two cases, FNAB was considered to be consistent with a benign tumour, but not diagnostic of pleomorphic salivary adenoma (Table 2).

Inflammatory lesions
Among the 34 cases of reactive or inflammatory lesions, FNAB was suggestive of reactive or inflammatory pathology and not suspicious for malignancy in 21. In seven cases, FNAB was considered either inconclusive or suspicious. In the other six cases, the samples were considered inadequate. There was no case of a FNAB being falsely positive for malignancy in a lump which subsequently turned out not to be malignant. Excluding the six cases with inadequate samples, therefore, FNAB was considered not to be suspicious for malignancy in 75% of cases of reactive or inflammatory swellings, and inconclusive in 25%. (Table 2).

Cystic lesions
Fine needle aspiration biopsy was suggestive of a benign pathology in six of the eight cases of branchial cleft cysts. Five of these cases showed typical bland squames suggestive of a diagnosis of a branchial cyst. In one case, the specimen obtained by FNAB was considered inadequate, while in one other, FNAB was considered inconclusive. Fine needle aspiration biopsy was suggestive of a benign pathology in the three cases of thyroglossal duct cyst and the single case of a dermoid cyst. (Table 2)

Of the eight cases of cystic lateral neck masses undergoing FNAB that were not considered to be suspicious for malignancy, six turned out to be branchial cleft cysts, and two cystic metastases, giving a false negative rate for malignancy among lateral neck cysts of 25%. This figure does not include four additional cases of cystic metastases that were diagnosed as such only after repeat aspiration.

DISCUSSION
Fine needle aspiration biopsy has become established as an investigation of choice in the diagnosis of neck lumps. It is minimally invasive, does not violate the neck, and is very accurate in most cases. In addition, it allows for the rapid diagnosis of malignant pathology and enables the initiation of prompt and appropriate treatment.

The accuracy of FNAB is dependent on obtaining an adequate specimen for examination by the cytopathologist. Nearly one fifth of all aspirates taken in the present series were considered inadequate. The correct technique for performing FNAB is described elsewhere.^3,6 The presence of a cytopathologist at the time that the procedure is being performed is considered to enhance the likelihood of obtaining an adequate specimen.⁶ When the sample obtained by FNAB is reported as being inadequate, then a repeat aspiration should be performed.

When inadequate samples are excluded, FNAB has a high sensitivity in the diagnosis of most neoplasms. False negatives may occur, however, so that in any case where there is a suspicion that a neck lump may be malignant, and where initial FNAB has been reported as benign, then repeat FNAB should be performed. Particular caution should be exercised in the case of cystic neck lumps. Most malignant neck lumps which were incorrectly diagnosed by FNAB in the present study were cystic, and 25% of cystic lateral neck lumps not considered to be suspicious for malignancy turned out to be malignant. This figure may have been higher if FNAB had not been repeated in four other cystic lumps (two with an inadequate initial aspirate, and two not considered to be suspicious for malignancy on initial FNAB). Repeat FNAB was diagnostic of malignancy in two of the three cases of cystic carcinoma, where repeat aspiration had been performed after initial FNAB had not been considered suspicious. On the other hand, only one aspirate had been performed in three of the four cases of cystic carcinoma with a false-negative FNAB. Had FNAB been repeated in these cases, then it is possible that the sensitivity of FNAB in the diagnosis of cystic carcinomas may have been improved.

The difficulty in distinguishing between benign cysts and cystic carcinomas on the basis of FNAB has also been documented by other authors.^8-9 A substantial rate of unsuspected malignancy in lateral cervical cysts has also been reported, particularly among adults over the age of 40 years.^8-11 We would, thus, suggest considering any cystic neck mass in an older adult as malignant until proven otherwise. When FNAB is negative for malignancy in such cases, then consideration should be given to repeating this procedure. Our experience would suggest that repeating the FNAB may lead to an increased sensitivity for FNAB in the detection of cystic carcinomas.

Lymphoma cannot be accurately diagnosed by FNAB. However, depending on the type of lymphoma, this diagnosis may be strongly suggested.3 In other cases, however, the aspirate may closely resemble that from a reactive lymph node.

Conversely, aspirates from reactive lymph nodes may closely resemble lymphoma.⁶ In the present series, over one quarter of patients with lymphoma showed no strong suggestion of this diagnosis on FNAB, while one quarter of patients with reactive or inflammatory lymph nodes had aspirates which were considered suspicious. We believe that open biopsy of a neck lump should be performed in any case where FNAB is suggestive of lymphoma or where the lump persists, even if FNAB is not suspicious. Violating the neck is not a prime concern in these cases, as it is unlikely for a solid neck lump that has already undergone FNAB to unexpectedly show squamous cell carcinoma after open biopsy.³ However, open biopsy inevitably carries the risks of damage to the marginal mandibular nerve and accessory nerve, as well as wound complications and scarring.

The work-up of a patient with a neck lump of unknown aetiology obviously begins with a full history and physical examination. Examination of the mucosal surfaces of the upper aerodigestive tract by flexible nasopharyngoscopy is essential as this may reveal an unsuspected primary tumour, as is examination of the salivary glands, thyroid gland, and the skin of the face, neck, and scalp, which may harbor an unexpected malignant melanoma.¹ With supraclavicular triangle masses, the possibility of a primary site below the clavicles should also be borne in mind. Fine needle aspiration biopsy is the next step. If this is positive for neoplasm, then further management is dictated by the clinical circumstances. If FNAB is suspicious for lymphoma, or if it is not suspicious but the lump is persistent, an open biopsy should be performed. In older patients or those with a history of heavy smoking and/or alcohol consumption, who may be at increased risk of squamous cell carcinoma and where initial FNAB is negative for malignancy, should be FNAB repeated. In such cases, consideration should also be given to performing panendoscopy with directed biopsies of areas known to have a high incidence of occult primary neoplasms (nasopharynx, tongue base, tonsil, and piriform fossa), prior to any open biopsy.¹² Consideration should also be given to obtaining consent for radical neck dissection pre-operatively, and to proceed with this if frozen section examination of the neck lump shows squamous carcinoma.^8-10

Our experience of FNAB confirms it to be a safe and effective procedure, with a high diagnostic accuracy for carcinomas, salivary gland tumours and other lesions. However, vigilance should be exercised in the case of cystic masses, which may show a high incidence of unsuspected carcinoma. The diagnostic accuracy of FNAB may be improved in these cases by having a low threshold for repeating the FNAB in cases where the initial aspiration has been negative for malignancy. The other potential pitfall concerns the distinction between reactive nodes and lymphoma. Although FNAB may be suggestive of a diagnosis of lymphoma (majority of cases) a substantial proportion of aspirates from lymphomas are likely not to be considered suspicious. However, the risk of carcinoma being present in a solid neck lump where FNAB has been negative, is low. Finally, a high index of suspicion should be maintained in any patient considered to be at particular risk of carcinoma on the basis of age, smoking status or other clinical symptoms (hoarseness, dysphagia, throat or mouth pain, otalgia, or weight loss). In such cases, repeating the FNAB may reduce the likelihood of missing a cancer, particularly in cases where initial sampling may have been inadequate.

REFERENCES
1. McGuirt WF. Differential diagnosis of neck masses. Cummings CW, Fredrickson JM, Harker LA, Krause CJ, Richardson MA, Schuller DE (Eds): Otolaryngology and Head and Neck Surgery. 4th ed. St. Louis: Mosby; 1998; 1686-99.
2. Slaughter DP, Marajakis JD, Southwick HW. Clinical evaluation of swellings in the neck. Surg Clin N Am 1956; 36: 3-9.
3. Patt BS, Schaefer SD, Vuitch F. Role of fine needle aspiration in the evaluation of neck masses. Med Clin N Am 1993;77:611-23.
4. McGuirt WF, McCabe BF. Significance of node biopsy before definitive treatment of cervical metastatic carcinoma. Laryngoscope 1978;88:594-97.
5. Royston D. Fine needle aspiration biopsy of lymph nodes and subcutaneous masses. Ir J Med Sc 1993;162:21-23.
6. Amedee RG, Dhurandhar NR. Fine needle aspiration biopsy. Laryngoscope 2001;111:1551-57.
7. Frable WJ, Frable MA. Thin-needle aspiration biopsy in the diagnosis of head and neck tumours. Laryngoscope 1974;84:1069-77.
8. Gourin CG, Johnson JT. Incidence of unsuspected metastases in lateral cervical cysts. Laryngoscope 2000;110:1637-41.
9. Flanagan PM, Roland NJ, Jones AS. Cervical node metastases presenting with features of branchial cysts. J Laryngol Otol 1994;108:1068-71.
10. Cinberg JZ, Silver CE, Molnar JJ, Vogl SE. Cervical cysts: cancer until proven otherwise? Laryngoscope 1982;92:27-30.
11. Granstrom G, Edstrom S. The relationship between cervical cysts and tonsillar carcinoma in adults. J Oral Maxillofac Surg 1989;47:16-20.
12. Leipzig B, Winter ML, Hokanson JA. Cervical nodal metastases of unknown origin. Laryngoscope 1981;91:593-98.

Copyright: 10 April 2004