Miriam E. Tucker
Nearly three quarters of dedicated thyroid ultrasounds don’t identify biopsy-recommended nodules, and over a third don’t identify any nodules, new research finds.
“The number of thyroid ultrasounds performed in the United States has increased fivefold since 2002. This substantial increase produces a significant strain on healthcare resources and leads to over-detection and overtreatment of benign thyroid nodules and small, indolent cancers with questionable clinical relevance,” wrote Elena Kennedy, MD, then a medical student in the department of surgery at the University of Wisconsin School of Medicine and Public Health, Madison, and colleagues.
The data, published online on December 7, 2023, in Thyroid, come from a retrospective chart analysis of more than 1700 people who underwent dedicated (ie, specifically to look for a nodule) thyroid ultrasounds at a tertiary academic center. The rates of detecting both nodules and biopsy-recommended nodules were highest when the indication was a nodule seen incidentally on other imaging (aka “incidentaloma”) and lowest when the ultrasound was ordered because the patient had either metabolic or compressive symptoms.
And for the most commonly listed indication, a suspected palpable nodule, nearly half of the ultrasounds found no nodule and only one in five detected a nodule that warranted a biopsy.
The principal investigator of the study David O. Francis, MD, an otolaryngologist at the University of Wisconsin, Madison, told Medscape Medical News, “Thyroid cancer has grown in incidence three to four times over the last 30 years without a good explanation for why…It seems to be that we’re detecting smaller and smaller nodules…Why are people being referred for all these ultrasounds? We looked for the upstream factors.”
One clear clinical implication of the new data, Francis noted, is that “if someone has compressive symptoms including dysphagia, swallowing problems, voice change, or globus sensation, ultrasound should not be the first way to work them up…It would be smarter to have someone evaluate their voice or their swallowing to see if there’s another reason besides the thyroid. The thyroid would have to get pretty big to cause dysphagia or swallowing problems.”
No Current Guidelines Advise When not to Order a Thyroid Ultrasound
Problematically, while there are professional society guidelines for what to do when a thyroid “incidentaloma” is found and other specific situations, there are no overall guidelines addressing when it’s appropriate to order a thyroid ultrasound, Kennedy, now an otolaryngology resident at the Indiana University of Indianapolis, and colleagues, point out.
According to Francis, “Ultrasounds are low cost and low risk. Those two factors result in people ordering more tests…The problem with that is we find things, and then we have to figure out what to do with them. That leads to incidentalomas, the surveillance, worry and anxiety, and costs…It’s tricky. We don’t want to discourage people from ever ordering ultrasounds, but there need to be some guidelines around when it’s appropriate to order.”
Asked to comment, Trevor E. Angell, MD, associate medical director of Thyroid Center at Keck School of Medicine of the University of Southern California, Los Angeles, told Medscape Medical News that the study is “clinically very important.”
Angell pointed out that the current American Thyroid Association (ATA) guidelines on thyroid nodule management, of which he is an author, recommend ultrasound for a known or suspected nodule. But he added, “there certainly should be a message that obtaining ultrasound for these other reasons are less likely to identify a nodule or anything causative. Whether it’s gastroesophageal reflux or allergic rhinitis or vocal cord dysfunction, an ultrasound isn’t a good test for those either.”
Angell said that the next ATA thyroid nodule guidelines, expected out in 2024, will address this topic more fully, but he couldn’t provide more specific information because the document is still in development. He did say, however, “Addressing when not to do an ultrasound will be an important consideration in the next guidelines.”
Low Detection Rates for Most Indications
The retrospective observational cohort study included 1739 adults (76% women; mean age, 53 years) who underwent dedicated thyroid ultrasounds between 2017 and 2019. In most cases, the recommendation for biopsy was determined using the American College of Radiology TI-RADS system, based on nodule size and TI-RADS category.
The most common indication for thyroid ultrasound, suspected palpable nodule, accounted for 40% of those performed. Follow-up for an “incidentaloma” was the indication in 28% of patients, and referral for compressive and metabolic symptoms accounted for 13% and 6% of ultrasounds, respectively.
Among all ultrasounds performed, 62% identified a thyroid nodule. Patients referred for incidental findings had the highest percentage of ultrasounds with thyroid nodules present at 94%. By contrast, in those referred for suspected palpable nodule on exam and for compressive symptoms, nodules were identified on 55% and 39% of ultrasounds, respectively. Patients with metabolic symptoms had a nodule identified on ultrasound 43% of the time. Among those referred for high risk factors, 57% had a nodule present.
Overall, only 27% of ultrasounds identified a thyroid nodule that was recommended for a biopsy. Again, those referred because of an incidental imaging finding had the highest percentage (55%), followed by those referred for a suspected palpable nodule (21%), high risk factors (20%), combined indications (16%), metabolic symptoms (10%), and compressive symptoms (6%).
Mean nodule size was largest among the patients referred for incidentalomas (2.4 cm), whereas all the other groups had mean nodule sizes between 1.2 cm and 1.8 cm, a significant difference (P < .05). The median size of nodules among those referred to ultrasound for a suspected palpable nodule was 1.4 cm.
“That’s pretty small. It would have had to be in the front of the thyroid where they could actually touch it. I would argue that the number of clinicians who actually palpated something was smaller. We’ve done several projects looking at how small a nodule a clinician can actually feel in the thyroid gland from the neck. It turns out we’re pretty bad at physical examination of the thyroid. This paper kind of reinforces that,” Francis told Medscape Medical News.
Patients with incidental nodules were over 10 times more likely to have a nodule found on an ultrasound than those referred for a suspected palpable nodule on exam (odds ratio [OR], 10.6). Conversely, those referred for compressive symptoms were half as likely to have an identifiable nodule compared with those referred for physical exam findings (OR, 0.5).
The odds of finding a nodule increased with age, especially for those aged ≥ 65 years compared with those younger than 45 years (OR, 3.6). Women were twice as likely to have a nodule found on thyroid ultrasound (OR, 2.0). Results were similar for the biopsy-recommended nodules, except that there was no difference between sexes (female vs male OR, 1.2).
Angell called the study “a very robust comprehensive evaluation,” but also noted that the single center source is a limitation. “It would be nice to have those big databases of national healthcare settings, but getting that granular level of information about why something was done is nearly impossible in that context.”
Kennedy, Francis, and Angell have no disclosures.
The study was published online on December 7, 2023, in Thyroid.
Miriam E. Tucker is a freelance journalist based in Washington DC. She is a regular contributor to Medscape, with other work appearing in the Washington Post, NPR’s Shots blog, and Diabetes Forecast magazine. She is on X (formerly known as Twitter) @MiriamETucker.
Leave a Reply