The Role of Tc-99 m-MIBI Scan in the Assessment of the Cystic Thyroid Nodule

Abstract

Purpose

A non-surgical therapeutic option requires assurance that a cystic thyroid nodule with non-diagnostic cytology is benign. This work was undertaken to determine whether Tc-99 m-MIBI scan (MIBI) findings can guide the best therapeutic option with confidence.

Material and Methods

We studied 81 cystic non-functioning thyroid nodules with non-diagnostic fine-needle aspiration biopsy (FNAB) report classified according to ATA 2015 ultrasonography (US) patterns for suspicion of malignancy. All had a MIBI to assess metabolic activity within the nodule as well as histopathological diagnosis. Diagnostic value analysis of MIBI as compared to the US pattern was determined.

Results

None of the 11 patients with US pattern of benign showed positive MIBI, and all had a histopathological report of benign. Diagnostic value of MIBI on US pattern of very low suspicion showed sensitivity, specificity, PPV, and NPV of 100%, 78.9%, 42.9%, and 100%, respectively.

Conclusion

Our data shows that the only approach to a safe non-surgical treatment option in a cystic nodule with non-diagnostic FNAB is when no metabolic activity is seen on MIBI.

Introduction

The main purpose for assessing a cystic thyroid nodule is to determine the best treatment option by differentiating benign from malignant lesions [1, 2].

Currently, according to the ATA 2015, there are three ultrasound (US) patterns for risk of malignancy involving cystic thyroid nodules: 1, benign pattern (0% risk of malignancy) with completely cystic nodules with well-defined walls; 2, very low suspicion pattern (< 3% risk of malignancy) with spongiform nodules and nodules with interspersed cystic spaces, without any of the features in more suspicious patterns; and 3, low suspicion pattern (5–10% risk of malignancy) characterized by isoechoic or hyperechoic nodule, partially cystic nodule with a peripheral solid component and none of the following: microcalcifications, irregular margins, extrathyroidal extension, or taller than wide [2].

In the cases of a benign US pattern on the assessment of thyroid cystic nodules, it is generally accepted that follow-up should be conservative monitoring. In the case of lesions with an US pattern of very low suspicion of malignancy, a fine-needle aspiration biopsy (FNAB) is suggested in those cases where lesions are > 2 cm in size, and in the case of lesions with an US pattern of low suspicion of malignancy, the FNAB is suggested where lesions are > 1.5 cm in size [1].

However, FNAB has an important limitation in the assessment of cystic lesions, due to the fact that the lesion content hardly contains any cells. Only in those cases where the cystic nodule has an identifiable lesion with solid content is that on occasions and with the help of US guidance it is possible to obtain a good FNAB for cytologic diagnosis. This explains why a large number of cystic lesions (with no solid identifiable area on US) in which a FNAB is done end up without a cytologic diagnosis (non-diagnostic or unsatisfactory) [3]. Importantly, when these cytology reports are reviewed, prevalence of malignant cells has been shown to be non-negligible [4].

There are several therapeutic options for the thyroid cystic nodule (e.g., conservative observation, FNAB [5], sclerosis [6], radiofrequency [7], laser [8], surgery [9]), and as most cystic lesions are benign, it is important to demonstrate with the best possible degree of certainty if the cystic lesion is benign or malignant, to justify and pursue the best therapeutic option, whenever possible, favoring the least invasive first.

The metabolic evaluation of non-functioning (as defined by no uptake of Tc-99mO₄⁻) thyroid nodules with Tc-99 m-methoxy-isobutyl-isonitrile (MIBI) has been widely studied [10–14]. When done with adequate technique [15–17], thyroid scans with MIBI showing no metabolic activity have been shown to have a high diagnostic value to rule out malignant lesions (negative predictive value of 100%); therefore, the best therapeutic options could be decided with confidence that no malignant lesion is being left to its natural evolution on those cases.

The objective of this work was to determine if the metabolic evaluation of the cystic thyroid nodule with MIBI brings enough information to decide the best therapeutic option with confidence.

Material and Methods

This is a retrospective, transversal, observational descriptive study on patients with non-functioning thyroid nodules with cystic content (≥ 80% of the whole nodule as determined by US), with non-diagnostic or unsatisfactory cytology report on FNAB and classified according to US patterns as described by the ATA guidelines for assessment thyroid nodules in 2015 [18].

Population

All procedures were performed as standard of care following the internal guidance of our institution for the assessment of patients with thyroid nodules, and the ethics committee and institutional review board approved this study. In our institution, all patients undergo a thyroid ultrasound as well as a FNAB when the FNAB fails to reach a diagnosis; complementary diagnostic work-up is routine clinical practice; therefore, all patients also underwent a functional thyroid scan with Tc-99mO₄⁻ as well as a metabolic scan with Tc-99 m-MIBI (MIBI). In addition to this, according to the surgical pathways at the Thyroid Clinic of the General Hospital of Mexico, as every thyroid nodule without a cytologic diagnosis is intervened, every patient had a thyroid surgery.

Image Acquisition

All planar thyroid scans (Tc-99mO₄⁻ as well as Tc-99 m-MIBI scans) were obtained in the anterior projection of the neck with a Siemens Orbiter 37 tube detector SPECT gamma camera connected to a dedicated Siemens ICON image acquisition and processing workstation as display terminal. The camera was equipped with an ultra-high resolution parallel-hole, low-energy collimator. Images were obtained in a 256 × 256 matrix using a digital zoom of 2. The acquisition time was set to 10 min with a 20% window centered at 140 keV in all cases.

Thyroid scintigraphy with Tc-99mO₄⁻ to assess tracer uptake (functionality) of the lesions was obtained as routine standard of care. Briefly, a 10-min-long anterior projection image of the thyroid is obtained at 20 min after an intravenous injection of 300–370 MBq of Tc-99mO₄⁻. The lesion of interest is subjectively assessed by an experienced radiologist as either a non-functioning or a functioning lesion and reassessed for confirmation by the first author of this study.

The MIBI scans were obtained always at least 2 days apart and as previously described [10]. Briefly, the MIBI scans were obtained at 15 and 120 min after the intravenous administration of 296–370 MBq of Tc-99 m-MIBI. In a binary way, metabolic activity within the cystic nodule was subjectively assessed by 2 physicians (first and second authors) with more than 10 years of experience assessing these tests. The MIBI scans were classified as either MIBI positive (activity superior to background and surrounding tissues seen within the lesion) if the early and/or the late scan showed any degree of MIBI uptake or as MIBI negative (activity equal or inferior to background activity in surrounding tissues seen within the lesion) when on both images no MIBI uptake was seen. All patients also underwent thyroid surgery and have a histopathological report as the final diagnostic gold standard.

Thyroid ultrasounds and FNAB were performed using a real-time ultrasound equipment (Resona 7, Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, PR China) with a 7.5 MHz linear array transducer and were done as standard of care by an institutional radiologist with more than 10 years of experience performing and reporting thyroid ultrasounds.

Statistical Analysis

Sensitivity, specificity, accuracy, and positive and negative predictive values (PPV and NPV, respectively) of MIBI were calculated as compared to the US pattern of cystic thyroid nodules.

Results

Out of 1045 patients with thyroid nodules seen at the Thyroid Clinic General Surgery Service of the General Hospital of Mexico between January 2017 and January 2019. Ninety-three (8.9%) patients (84 women and 9 males, with mean age of 43.1 years with mode of 55 years and range of 17–63 years) were diagnosed by US as having a cystic nodular thyroid lesion, a partially cystic lesion or a partially cystic with eccentric solid lesion.

FNAB was diagnostic in only 12 (12.9%) of the 93 patients, all with cytology report of benign (Bethesda II) in samples obtained from an eccentric solid area. In the other 81 patients, the cytology reports only determined “cystic lesion” and were not able to rule in or rule out the presence of malignancy; these are the 81 cases that were further analyzed for the purpose of this paper.

An US pattern of benign was seen in 11 (13.6%) nodules with a size of 16 mm (± 0.39 mm SD), a very low suspicion pattern in 22 (27.1%) nodules with a size of 17 mm (± 0.45 mm SD), and a low suspicion pattern in 48 (59.2%) nodules with a size of 19 mm (± 0.57 mm SD).

MIBI scans were negative in the nodules of 60 (74%) patients (malignancy was ruled out by histopathology on surgical specimen on all of these), and MIBI was positive in 21 (25.9%) patients, of which 10 (47.6%) had a histopathological report of malignancy (papillary thyroid carcinoma). There were no discrepancies between the readers regarding the functionality of the lesion or the assessment of the early and late MIBI scans. Additionally, all positive MIBI scans at 15 min were also positive 120 min later, and all negative MIBI scans at 15 min were also negative 120 min later.

None of the 11 patients with US pattern of benign had a positive MIBI, and all 11 had a final histopathological report of benign (Fig. 1). Of the 22 lesions with an US pattern of very low suspicion, 7 (31.8%) had a positive MIBI, and the final histopathological assessment found papillary thyroid carcinomas in 3 of these patients, all with single papillary thyroid carcinoma of 5.6 and 9 mm in maximum length present in the thyroid nodule studied and staged according to TNM as T1aN0M0. Finally, of the 48 lesions with an US pattern of low suspicion, 14 had a positive MIBI (Fig. 2), and the final histopathological assessment found papillary thyroid carcinomas in 7 patients, all with single papillary thyroid carcinoma of 5, 6, 8, 9, 11, 14, and 15 mm in maximum length present in the thyroid nodule studied; four cases were staged as T1aN0M0 and three as T1bN0M0 according to TNM.

Fig. 1.

Non-functional right cystic thyroid nodule with benign ultrasonographic pattern, negative Tc-99mO4- and negative MIBI. Histopathology was benign

Fig. 2.

Non-functional left cystic thyroid nodule with low suspicion pattern of malignancy on ultrasonography, negative Tc-99mO4-, and positive MIBI. Histopathology revealed cancer

The diagnostic value of MIBI on those thyroid cystic nodules with a very low suspicion pattern for malignancy by US showed sensitivity, specificity, PPV, and NPV of 100%, 78.9%, 42.9%, and 100%, respectively.

Likewise, the diagnostic value of MIBI on those thyroid cystic nodules with a low suspicion pattern for malignancy by US showed sensitivity, specificity, PPV, and NPV of 100%, 82.9%, 50% and 100%, respectively (Table 1).

Table 1.

Positive predictive value = PPV. Negative predictive value = NPV. CI = confidence interval

	MIBI negative cancer present	MIBI negative no cancer	MIBI positive cancer present	MIBI positive no cancer	Sensitivity % (CI 95%)	Specificity % (CI 95%)	PPV % (CI 95%)	NPV % (CI 95%)
US ATA low pattern	0	34	7	7	100 (64.6–100)	82.9 (68.7–91.5)	50 (26.8–73.2)	100 (89.8–100)
US ATA very low pattern	0	15	3	4	100 (43.8–100)	78.9 (56.7–91.5)	42.9 (15.8–75.0)	100 (79.6–100)

Discussion

When FNAB fails to reach a diagnosis, performing thyroid scintigraphy and a MIBI scan is routine clinical practice in our institution. Our results show once again that in the assessment of the non-functioning thyroid nodule (as assessed by scintigraphy), when there is no uptake of MIBI (no metabolic activity), it is safe to assume that the lesion is benign (NPV 100%).

Therefore, all non-functioning cystic thyroid nodules with a non-diagnostic FNAB should be evaluated with a MIBI scan; if there is no MIBI uptake (no metabolic activity), the thyroid nodule can be considered as benign with 100% certainty.

On the other hand, if a thyroid nodule has a non-diagnostic FNAB but there is MIBI uptake in the nodule, further action must be taken to try to reach a more definitive diagnosis (benign vs malignant lesion) in order to consider well-informed best treatment options, as a positive MIBI scan cannot used to rule in or to rule out malignancy (high false positive rate). In our group, we consider that this action should be a diagnostic thyroid surgery, since on those patients with an US showing a pattern of very low suspicion (with an ultrasonographic probability of malignancy of ≤ 3%), the probability increases to 13.6% with a positive MIBI. Furthermore, on those patients with an US showing a pattern of low suspicion, the probability of malignancy increases from 10 to 14.5% with a positive MIBI.

Although it is true that ultrasound has some capacity to identify nodules with a low risk of having cancer (except in the “benign” category where it has a 100% capacity as our results corroborate), it is not possible to establish this with a diagnostic capacity of 100% in the “very low” and “low” categories, since in our series we found a 3 to 10% probability of malignancy, respectively. However, MIBI scintigraphy has this possibility, and given that most of the type of nodules studied fall into these two categories (“very low” and “low”), undoubtedly the only way to offer absolute certainty to “rule out” cancer would be to carry out a thyroid MIBI scan and demonstrate that said nodule has no metabolic activity.

Although to the best of our knowledge no previous work studying exclusively cystic thyroid nodules has been published, the concept of ruling out malignancy in thyroid nodules with MIBI scintigraphy is not new and has been reported by several groups in the past [10–17]. Therefore, the authors believe that it is important that when a physician decides to do a non-surgical treatment of a thyroid nodule, he has the complete certainty of having ruled out malignancy. This work shows that MIBI scans help to achieve this purpose.

Newer technologies (e.g., SPECT/CT and PET imaging) [19] with better resolution and quantitative possibilities could potentially improve the positive predictive value of the MIBI scans (which have been shown to have the limitation of a high false positive rate) or other radiopharmaceutical to positively identify malignant thyroid lesions, but so far, this has not been done. We believe that more studies should be undertaken to find a diagnostic test that can accurately differentiate between all malignant and benign thyroid lesions.

Conclusion

The need to characterize the thyroid nodules with indeterminate FNA for therapeutical decisions is well recognized. We can conclude that because of the negative predictive value of the MIBI scans (100%), the only approach to a definitive safe non-surgical treatment option in a patient with a non-functioning thyroid cystic nodule with non-diagnostic FNAB can be taken only on those patients with a negative MIBI scan (no uptake, no metabolic activity) within the cystic lesion. Moreover, due to the low positive predictive value of the MIBI scans (high false positive rate) on all cases where there is MIBI uptake (metabolic activity) in the lesion, we recommend following diagnostic and therapeutic institutional guidance, including the surgical approach.

Author Contribution

Luis-Mauricio Hurtado-López: conceptualization, methodology, investigation, and writing — original draft. Carlos Martinez-Duncker: conceptualization, methodology, investigation, and writing — original draft.

Data Availability

The data in this study come from the archives of the thyroid clinic of the General Hospital of Mexico and can be shared for your review. Please contact author for data requests.

Declarations

Competing Interests

Luis-Mauricio Hurtado-López and Carlos Martínez-Duncker declare no competing interests.

Ethics Approval and Consent to Participate

All procedures were performed as standard of care following the internal guidelines of the institution for the assessment of patients with thyroid nodules, and the IRB and the ethics committee approved this study.

Consent for Publication

Our work does not require informed consent of images as it does not contain any individual person’s data in any form (including individual details, images, or videos).