Abstract
C
Given our acquired successful experience of active surveillance (AS) for patients with Bethesda V/VI thyroid nodules, 4 which have a higher risk of malignancy (60–95%), we considered a similar strategy with follow-up for nodules classified as Bethesda IV. This strategy would offer the opportunity for selective surgical management based on ultrasound findings, cytological characteristics, and clinical behavior, and reduce unnecessary diagnostic lobectomies. To date, there are few studies evaluating AS in patients with Bethesda IV nodules. 5 This study describes the evolution and clinical behavior of patients with Bethesda IV thyroid nodules who were referred to surgical treatment but underwent AS.
This is a retrospective analysis of a cohort recruited from January 1, 2016 to December 1, 2021 at a surgical Head and Neck Cancer center in Medellin, Colombia. This study was approved by the Research Committee of the Faculty of Medicine of the University of Antioquia (IRB ID 00012257). In accordance with the Declaration of Helsinki, written informed consent was waived by the ethics board due to the descriptive nature of the study. All patients were referred by clinicians for consideration of surgery and the cases were reviewed by staff surgeons.
Patients older than 18 years, with no prior history of head and neck radiation exposure nor any family history of thyroid cancer, who had intrathyroidal nodule <4 cm on ultrasound and a cytological diagnosis of Bethesda IV were considered candidates for AS. Patients with previous head and neck cancer, high suspicion of thyroid malignancy, or multinodular goiter with surgical indication were excluded. We included only patients who had at least 12 months of follow-up in this report.
We followed the protocol previously defined by our team for AS of patients with Bethesda V–VI nodules. 4 Patients underwent US every 6 months during the first year and annually thereafter. Surgery was recommended if there was an increase in the largest diameter >3 mm, ultrasonographic evidence of high-risk characteristics (ill-defined borders, suspicious lymph nodes, and microcalcifications), or surgeon/patient decision to stop surveillance.
The medical records were reviewed, and data about age, sex, cytology results, ultrasound characteristics (American College of Radiologists-Thyroid Imaging Reporting and Data Systems [ACR-TIRADS]), surgical indication, type of surgery, and follow-up results were collected.
The time from the fine needle aspiration biopsy result to the last follow-up visit or surgery was calculated. A Kaplan–Meier graph was designed to evaluate time-to-event outcomes (Stata 14.0. Stata Corp).
Fifty-three patients with a Bethesda IV thyroid nodule diagnosis were analyzed. We excluded two patients who had follow-up for <12 months. The mean age was 50.9 ± 13.8 years (range 20–76), 21 patients (40%) were older than 55 years and 45 patients (84.9%) were female. The mean nodule size was 15.4 ± 4.9 mm (median 15 mm, 5–29 mm). Seven (13.2%) patients had nodules ≤10 mm in diameter, 39 (73.6%) nodules 11–20 mm in diameter, and 7 (13.2%) nodules >20 mm in diameter. The distribution of the TIRADS classification of the nodules was as follows: TIRADS-1, 7 (13%); TIRADS-2, 21 (40%); TIRADS-3, 11 (21%); and TIRADS-4, 14 (26%). Findings of Hurthle cells in fine needle aspiration biopsy (FNAB) smears occurred in nine patients (18.3%).
The median and mean follow-up periods were 36.4 and 37.7 months, respectively (6.1–81.8 months). Forty-nine patients had a follow-up longer than 18 months. Twenty-two (41.5%) patients had nodules that remained stable or decreased in size. Nodule growth >3 mm in diameter occurred in 16 patients (30.2%); 2 in nodules ≤10 mm, 12 in nodules 10–20 mm, and 2 in nodules >20 mm. Ten (18.8%) patients underwent surgery: eight cases due to growth (mean 13.4 mm) and two cases due to the surgeon's decision.
Of these 10 patients, five cases were reported as intermediate-risk papillary carcinoma due to microscopic extrathyroidal extension, one patient had an incidental finding of low-risk carcinoma in a nonindex nodule (the index nodule was a follicular adenoma), and the others were reported as adenoma. Of the eight patients who had a growth >3 mm who were not operated on, one case had a growth of 17 mm, and an FNAB was repeated with a result of Bethesda II. In the others, the surgeon opted to continue surveillance (mean change 5.7 mm). None of the patients under AS developed nodal/distant metastases or died.
Follow-up without surgery at 24 months (Fig. 1) was 90.3% (95% CI 0.78–0.95). None of the 10 patients reported with Hurthle cells underwent surgery, and only 3 patients who continued AS showed growth (2.6 mm) during observation.
Surgery-free follow-up in patients with Bethesda IV thyroid nodules.
Smulever and Pitoia 5 have previously discussed that in developing countries, access to molecular testing for indeterminate cytology thyroid nodules is limited (as it is not covered by the health system), so the only viable alternative for most patients is surgery. As detection of thyroid nodules has increased in recent decades due to the higher availability of ultrasound, a larger number of thyroid biopsies are being performed. It is expected that almost 10–20% of them will be finally reported as Bethesda IV, and these patients could undergo a lobectomy. 2 In a recent series of 1,379 patients with follicular neoplasia, 34% were diagnosed with malignancies, and none of these was a follicular thyroid carcinoma. 6 However, almost 60% of patients with Bethesda IV nodules will not obtain any benefit from surgery and are at risk of surgical complications (1–3%) and likelihood of long-term levothyroxine support (10–48%). 3 In this cohort, we did not find any cases of follicular carcinoma.
In most health systems, most patients are primarily evaluated by clinicians and are referred to surgical treatment based in the cytological result. This situation encouraged us to evaluate a strategy of AS adjusted by TIRADS classification. We based our approach on the moderate diagnostic performance of ultrasound to define malignancy and several proposals to integrate ultrasound and cytological findings with individual risk characteristics. 7 It can be inferred that these nodules may have a better prognosis than Bethesda V/VI nodules, and these patients are candidates for a more conservative approach, such as AS. Although we only use ultrasound characteristics as size change to indicate surgery in the follow-up, repeating the FNAB could be a useful alternative in nodules that grow but maintain ultrasonographic low-risk characteristics.
Our population was similar to that reported by other studies (women with an incidental thyroid nodule <2 cm) and with low or intermediate risk according to TIRADS criteria. In the present cohort, 41% of nodules remained stable and only 16 patients had significant growth. At 24 months, 90% of patients had not undergone surgery, similar to the results of AS of nodules with a higher risk of malignancy. Of those operated, the rate of malignancy was 50% (5/10).
To date, active conservative management of Bethesda IV nodules has been proposed for patients with comorbidities, high surgical risk, or those who do not accept surgical treatment. 8 There are few studies assessing this strategy. 9 Smulever and Pitoia 5 reported 23 patients with Bethesda IV nodules who had high surgical risk or rejected surgery. They found a rate of growth >3 mm in three patients (14%) after 42 months of mean follow-up, and only one of these three patients underwent surgery with a histopathological report of benignity. After an extensive literature search, our study had the largest number of patients with Bethesda IV nodules under AS.
This study has some limitations. The sample size was small, and the follow-up time was still too short to make general recommendations. Observer bias due to interobserver variability of pathologists, radiologists, as well as selection bias by surgeons limit the generalizability of results.
In conclusion, an AS strategy in conjunction with an individual evaluation of US and cytopathological findings in selected patients with Bethesda IV thyroid nodules (intrathyroidal nodules found incidentally and without high-risk ultrasonographic characteristics) is feasible and may be an alternative to initial diagnostic lobectomy when molecular tests are unavailable, with rigorous adherence to the follow-up. Additional studies with a greater number of patients and extended follow-up periods are required to use this strategy widely in clinical settings.
Footnotes
Authors' Contributions
Conceptualization, methodology, validation, investigation, data curation, writing—original draft, and writing—review and editing by C.G. Conceptualization, methodology, validation, investigation, data curation, formal analysis, writing—original draft, and writing—review and editing by A.S.