False Negative Rate of Thyroid FNA
The false negative rate of thyroid FNA ranges from approximately 5-7% in most contemporary series, though rates as high as 11-15% have been reported for specific nodule characteristics, particularly subcentimeter nodules and large cystic/solid lesions. 1, 2
Overall False Negative Rates from High-Quality Studies
- A large 10-year institutional study of 695 thyroid nodules found an overall false negative rate of 7.0% for thyroid FNA 1
- A retrospective analysis of 802 benign FNA cases with surgical follow-up demonstrated a 6% false negative rate when defined as interpretation error and inadequate sampling of malignant nodules 2
- When excluding noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP), the false negative rate decreased to 5% 2
- When micropapillary thyroid carcinomas were also excluded, the false negative rate further decreased to 3% 2
- Historical literature reports false negative rates ranging from 0.4% to 13% across various institutions 1
Critical Nodule Characteristics That Increase False Negative Risk
Nodule Size Effects
- Subcentimeter nodules (0.5-0.9 cm) have the highest false negative rate at 15.8%, significantly higher than larger nodules 1
- Nodules 1.0-3.9 cm demonstrate a false negative rate of 6.3% 1
- Large nodules ≥4 cm show a false negative rate of 7.1%, which is not significantly different from medium-sized nodules 1
Cystic/Solid Composition
- Large cystic/solid nodules (≥3 cm with mixed composition) carry the highest false negative rate at 30% 3
- Cystic/solid nodules of any size have a false negative rate of 25% 3
- Large nodules (≥3 cm) regardless of composition show a 17% false negative rate 3
- Solid nodules <3 cm have the lowest false negative rate at 0% in one series 3
Mechanisms of False Negative Results
Sampling Error vs. Interpretation Error
- Among false negative cases, 57% are due to sampling error (needle missing the malignant area) and 43% are due to interpretation error (pathologist misreading the cytology) 2
- Interpretation errors occur more frequently in follicular-patterned neoplasms (75%), while sampling errors are more common in non-follicular variants of papillary thyroid carcinoma (61%) 2
- Tumors with interpretation errors are significantly larger (mean 2.3 cm) compared to those with sampling errors (mean 1.4 cm) 2
Imaging-Cytology Discordance as a Red Flag
- When thyroid nodules show discordance between suspicious ultrasound features and benign FNA results, the malignancy rate increases dramatically to 23.3% 4
- Nodules with suspicious ultrasound features but benign cytology have an 11.4% malignancy rate 4
- Nodules without suspicious ultrasound features and benign cytology have only a 0.5% malignancy rate 4
Clinical Implications and Guideline Recommendations
- The NCCN explicitly warns that "although FNA is a very sensitive test, particularly for papillary, false-negative results are sometimes obtained; therefore, a reassuring FNA should not override concerns in the presence of worrisome clinical findings" 5
- Expert pathology review at the treating institution is critical, as FNA sensitivity varies by cancer type and institutional expertise 5, 6
- Medullary carcinoma may require additional immunohistochemical studies (calcitonin staining) to confirm diagnosis, as it can be missed on routine cytology 5
High-Risk Scenarios Requiring Heightened Suspicion
- Large (≥3 cm) AND cystic/solid nodules warrant strong consideration for diagnostic thyroid lobectomy even with benign FNA due to the 30% false negative rate 3
- Subcentimeter nodules require increased awareness of potential sampling error, with a false negative rate approaching 16% 1
- Nodules with suspicious ultrasound features (microcalcifications, irregular borders, central hypervascularity) showing imaging-cytology discordance should undergo repeat FNA 5, 4
- Clinical features suggesting malignancy (firm texture, fixation to structures, rapid growth, vocal cord paralysis, suspicious lymph nodes) should override benign cytology 5