Proposed Article Structure
1. Clinical context and rationale for TI-RADS adoption. 2. Study design and real-world setting. 3. Main utilization outcomes: ultrasound, FNA, and surgery. 4. Oncologic outcomes and disease severity at diagnosis. 5. Cost implications and adherence analysis. 6. Clinical interpretation, limitations, and implications for practice. 7. Conclusion and references.
Highlight
Implementation of ACR TI-RADS at a tertiary institution was associated with fewer thyroid ultrasounds and a higher proportion of patients undergoing only one ultrasound examination.
Despite this reduction in imaging follow-up, the number of thyroid fine-needle aspirations and surgeries did not decline after TI-RADS adoption.
Rates of surgically confirmed malignancy, tumor size, and stage were similar before and after implementation, suggesting no obvious shift in cancer detection severity.
Total healthcare costs were higher in the post-TI-RADS era, and adherence to ACR TI-RADS recommendations in a sampled post-implementation cohort was approximately 70%.
Background
Thyroid nodules are extremely common in clinical practice, with detection rates rising because of widespread imaging, increased health system access, and incidental findings on cross-sectional studies. Most nodules are benign, but the central clinical challenge is to identify the minority harboring malignancy while limiting unnecessary procedures, repeat imaging, patient anxiety, and avoidable costs.
Ultrasound-based risk stratification systems were developed to standardize reporting and guide biopsy decisions. Among these, the American College of Radiology Thyroid Imaging Reporting and Data System, or ACR TI-RADS, is one of the most widely adopted frameworks. It assigns points based on ultrasound features including composition, echogenicity, shape, margins, and echogenic foci, then links the resulting category to size-based recommendations for follow-up or fine-needle aspiration. Its stated goal is to improve consistency and reduce unnecessary FNA of low-risk nodules.
Retrospective derivation and validation studies have generally suggested that ACR TI-RADS can decrease biopsy volume without substantially compromising cancer detection. However, performance in tightly curated retrospective datasets does not always translate into routine care. Real-world implementation depends on reporting behavior, clinician acceptance, patient preference, local pathways, endocrine and surgical referral culture, and adherence to recommendations. The study by Zhao and colleagues is therefore clinically relevant because it evaluates institution-level outcomes before and after TI-RADS implementation rather than only theoretical test characteristics.
Study Design
Zhao et al conducted a retrospective observational study at a tertiary institution where ACR TI-RADS was adopted in 2018. The investigators compared two real-world eras: a pre-TIRADS period from 2014 to 2017 and a post-TIRADS period from 2019 to 2022. Excluding the implementation year is methodologically sensible because it avoids contamination from partial uptake and transitional practice patterns.
The full cohort included 13,156 patients undergoing thyroid ultrasound. Patients were grouped according to whether imaging occurred before or after implementation. The study examined utilization outcomes, including the number of thyroid ultrasounds, the number of ultrasounds per patient, FNA rates, and surgical rates. It also assessed pathological outcomes among surgical cases, including malignancy rate, tumor size, and stage. In addition, the authors performed a cost analysis covering these procedures.
A subset of patients in the post-TIRADS era underwent more granular review for ultrasound findings and adherence to ACR TI-RADS recommendations. This component is important because implementation studies can fail not because the system itself is ineffective, but because real-world compliance is incomplete or variable across radiologists and referring clinicians.
Key Findings
Imaging utilization decreased after ACR TI-RADS implementation
The most consistent signal in this study was reduced ultrasound utilization after TI-RADS adoption. The total number of thyroid ultrasounds fell in the post-TIRADS era, with 9,112 examinations compared with 10,811 in the pre-TIRADS era. In parallel, a larger proportion of patients in the post-TIRADS group had only one ultrasound performed: 70.8% versus 62.1%, with p less than 0.001.
Clinically, this suggests that TI-RADS may have its greatest immediate effect on surveillance behavior rather than on downstream invasive procedures. This is plausible because structured size- and feature-based follow-up thresholds can reassure clinicians that many low-risk nodules do not require repeated short-interval imaging.
Fine-needle aspiration and surgery did not decline
Although ultrasound utilization decreased, both eras had a similar number of FNAs and surgeries. This is the central finding of the paper and the one most likely to influence practice discussions. If the principal promise of TI-RADS is reduction in unnecessary biopsy, these institutional data suggest that such benefit did not materialize at the system level in this setting.
There are several possible explanations. First, patients referred for biopsy or endocrine/surgical review may already represent a higher-risk subgroup than general ultrasound populations. Second, clinicians may still proceed with FNA despite TI-RADS recommendations because of symptoms, growth history, prior external imaging, family history, radiation exposure, patient anxiety, multinodular complexity, or discordance with other guideline frameworks. Third, once a nodule enters a specialty clinic pathway, intervention thresholds may be influenced by local practice norms more than by radiology report language alone.
Cancer-related outcomes were unchanged
The surgically proven malignancy rate was similar before and after implementation. Median tumor size and stage were also similar across the two periods. From a patient-safety standpoint, this is reassuring in one sense: reducing follow-up ultrasound did not appear to result in detection of larger or more advanced cancers among surgical cases. At the same time, the absence of change in cancer yield underscores that lower imaging intensity alone did not reshape biopsy or operative selection enough to alter oncologic outcomes.
The findings therefore argue against both extreme interpretations. TI-RADS implementation was neither clearly harmful in terms of cancer stage migration nor clearly effective in lowering invasive procedure burden in this institutional experience.
Costs increased rather than decreased
Total healthcare costs for the measured procedures were higher in the post-TIRADS group at $4,772,158.00 compared with $4,280,081.98 in the pre-TIRADS group. On face value, this may appear counterintuitive because ultrasound utilization fell. However, overall costs are driven by multiple components, including biopsy, surgery, pathology, changes in unit pricing over time, perioperative care, and case mix. If the rates of FNA and surgery remain stable, savings from fewer ultrasounds may be too small to offset inflationary pressures or procedural costs.
This result is especially important for health systems and policymakers. A reporting framework that reduces imaging frequency does not necessarily generate net savings unless it also changes downstream intervention patterns. Economic claims for TI-RADS therefore require validation in contemporary local practice rather than extrapolation from diagnostic modeling studies.
Adherence to ACR TI-RADS recommendations was incomplete
In the subset analysis, adherence to ACR TI-RADS recommendations was observed in 444 of 635 patients, corresponding to 69.9% with a 95% confidence interval of 66.2% to 73.4%. This is a useful real-world estimate. It indicates that approximately 3 in 10 patients were managed in a way that did not align with the framework.
That level of adherence is respectable for a complex practice change, but it may be insufficient to achieve the reductions in FNA predicted by retrospective simulations. Importantly, nonadherence does not automatically mean inappropriate care. Clinicians may have had legitimate reasons for deviating from the algorithm. Still, from an implementation science perspective, incomplete uptake is likely one major explanation for the gap between expected and observed benefit.
Clinical Interpretation
This study addresses a question that matters more than diagnostic elegance: what happens after a structured reporting system is deployed in routine care? The answer here is nuanced. ACR TI-RADS appears capable of reducing repeat thyroid ultrasound, but that effect alone did not translate into fewer biopsies, fewer operations, or lower costs at this tertiary center.
For radiologists, the message is that structured reporting may improve surveillance stewardship, but the report is only one part of the care pathway. The impact of TI-RADS depends on how endocrinologists, otolaryngologists, surgeons, primary care clinicians, and patients act on the recommendations. For endocrinologists and surgeons, the findings highlight that invasive procedure rates are shaped by clinical context beyond sonographic category alone. For administrators, the data caution against assuming that TI-RADS implementation will automatically reduce expenditure.
The study also illustrates a broader principle in medical quality improvement: diagnostic classification systems often perform differently in real-world ecosystems than in retrospective research cohorts. Factors such as institutional referral bias, local cancer prevalence, access to biopsy, patient expectations, and medicolegal concerns all influence outcomes.
How These Findings Fit With Existing Evidence
The original ACR TI-RADS framework was designed to standardize interpretation and reduce unnecessary biopsies compared with less selective approaches. Prior studies have generally supported favorable diagnostic efficiency, although the balance between sensitivity and biopsy reduction varies by population. Comparative work has also shown meaningful differences between ACR TI-RADS, the American Thyroid Association ultrasound patterns, and other systems such as EU-TIRADS and K-TIRADS.
What Zhao et al add is implementation-level evidence rather than purely test-level evidence. Their findings align with the idea that structured systems can streamline follow-up imaging, yet they challenge assumptions that this will reliably reduce FNA volume in practice. This distinction is important. A risk stratification tool can be diagnostically sound and still yield limited operational benefit if adoption is partial or if downstream decision-making is driven by other factors.
Strengths and Limitations
Strengths
The study is notable for its large sample size of 13,156 patients and its evaluation of real-world outcomes across two multi-year periods. It examines not only imaging and biopsy utilization but also surgery, pathology, and cost, which provides a more clinically meaningful picture than reporting biopsy reduction alone. The adherence subset analysis is another strength because it helps explain why institutional outcomes may diverge from theoretical expectations.
Limitations
As a retrospective single-center study, the analysis is vulnerable to confounding by secular trends. Changes in referral patterns, clinician staffing, local policy, patient case mix, and healthcare pricing between 2014 and 2022 may have influenced outcomes independently of TI-RADS. The COVID-19 era, which overlaps the post-TIRADS period, may also have altered imaging utilization and care-seeking behavior.
The abstract does not provide adjusted analyses, absolute counts for FNA and surgery, malignancy subtype details, or granular reasons for nonadherence. Without these data, it is difficult to determine whether unchanged procedure rates reflect appropriate selection, clinician override of low-risk recommendations, or concentration of higher-risk referrals. The cost comparison is informative but may be affected by inflation and changes in billing structures over time. Finally, because this is a tertiary institution, generalizability to community settings may be limited.
Practical Implications for Clinicians
First, clinicians should not assume that implementing ACR TI-RADS alone will reduce biopsy rates. If reducing unnecessary FNA is a local priority, institutions may need complementary interventions such as standardized endocrine referral pathways, multidisciplinary review of discordant cases, audit-and-feedback systems, and education on when deviation from TI-RADS is justified.
Second, the reduction in repeat ultrasound is meaningful. Surveillance itself carries cost, inconvenience, and psychological burden. A more selective approach to follow-up may therefore still represent a worthwhile quality improvement even if biopsy rates are unchanged.
Third, adherence should be measured, not presumed. This study’s roughly 70% adherence rate suggests that implementation is an active process. Institutions adopting TI-RADS may benefit from tracking structured reporting completeness, concordance between recommendations and management, and reasons for override.
Fourth, outcome evaluation should go beyond biopsy counts. Cancer stage, tumor size, false reassurance concerns, patient-reported anxiety, and economic consequences all matter. A narrower focus on FNA volume may miss the real effects, good or bad, of risk stratification systems.
Conclusion
In this large retrospective single-center study, implementation of ACR TI-RADS was associated with fewer thyroid ultrasound examinations and less repeat imaging per patient, but it did not reduce FNA or surgery rates and did not lower overall healthcare costs. Surgically confirmed malignancy rate, tumor size, and stage remained similar across eras, suggesting no obvious worsening of oncologic outcomes. The observed adherence rate of 69.9% provides a likely explanation for why real-world results may differ from retrospective projections.
For clinical practice, the study supports ACR TI-RADS as a tool that may curb follow-up imaging, but it also shows that reporting reform alone may be insufficient to change downstream intervention patterns. Future multicenter studies should examine how local referral ecology, clinician behavior, patient preference, and implementation fidelity modify the real-world impact of thyroid nodule risk stratification.
Funding and ClinicalTrials.gov
No funding information or ClinicalTrials.gov registration number is provided in the source abstract. The study was retrospective and observational.
References
Zhao Y, Yew KX, See JY, Tan IE, Venkatanarasimha N, Too CW, Wong YW, Koh HS, Yuen HW, Ang KA, Au MKH, Teo CEH, Kiong KL. Real-World Outcomes Following the Implementation of ACR TI-RADS for Evaluation of Thyroid Nodules. The Laryngoscope. 2026-02-22;136(5):2370-2378. PMID: 41725027. URL: https://pubmed.ncbi.nlm.nih.gov/41725027/
Tessler FN, Middleton WD, Grant EG, Hoang JK, Berland LL, Teefey SA, Cronan JJ, Beland MD, Desser TS, Frates MC, Hammers LW, Hamper UM, Langer JE, Reading CC, Scoutt LM, Stavros AT. ACR Thyroid Imaging, Reporting and Data System (TI-RADS): White Paper of the ACR TI-RADS Committee. Journal of the American College of Radiology. 2017;14(5):587-595. PMID: 28372962.
Haugen BR, Alexander EK, Bible KC, Doherty GM, Mandel SJ, Nikiforov YE, Pacini F, Randolph GW, Sawka AM, Schlumberger M, Schuff KG, Sherman SI, Sosa JA, Steward DL, Tuttle RM, Wartofsky L. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016;26(1):1-133. PMID: 26462967.
Grani G, Lamartina L, Ascoli V, Bosco D, Biffoni M, Giacomelli L, Maranghi M, Falcone R, Ramundo V, Cantisani V, Filetti S, Durante C. Reducing the Number of Unnecessary Thyroid Biopsies While Improving Diagnostic Accuracy: Toward the “Right” TIRADS. Journal of Clinical Endocrinology and Metabolism. 2019;104(1):95-102. PMID: 30272188.
