Although neck palpation and thyroid ultrasound carry very low risk, a suspicious screening result can set off a chain of events that may lead to opportunities for harms. The next step after detection of a suspicious nodule is diagnostic evaluation with fine-needle aspiration of the lesion. The risks of thyroid fine-needle aspiration are hospitalization, postprocedural hematoma, and needle tract tumor implantation, although two observational studies suggest that the rate of each of the three outcomes is lower than 1%. More importantly, the results of cytology can lead to additional tests and surgery. In a meta-analysis of 23,445 nodules that were biopsied, 60% of the nodules were benign, and 5% of the nodules were malignant. However, the remaining 35% of the nodules required repeat biopsy or surgery. In patients who had a diagnostic lobectomy or excision, 64% of the nodules were benign on final histological evaluation.
Thyroid surgery for benign disease has the same risks as it does for malignancy. In 2013, about 80% of surgeries for small localized papillary thyroid cancers were total thyroidectomies and 20% were lobectomies. However, the 25-year cumulative risk of death caused by thyroid cancer does not differ by type of surgery. Surgical risks are lower for a lobectomy than for total thyroidectomy. In addition to the general risks of surgery, specific risks of thyroid surgery include recurrent laryngeal nerve injury and hypoparathyroidism. Recurrent laryngeal nerve injury causes vocal cord paresis, which can result in difficulty speaking, difficulty swallowing, and hoarseness. Breathing difficulties are possible if both laryngeal nerves are affected. Hypoparathyroidism leads to hypocalcemia. A cross-sectional population-based survey included 2,632 thyroid cancer patients, 2 to 4 years after diagnosis, from two U.S. SEER registries. The study reported that 25.8% of patients had voice changes more than 3 months after surgery, and 4.7% of patients had vocal fold paralysis or paresis.
A meta-analysis of hypoparathyroidism lasting more than 6 months after thyroidectomy produced a summary event measure of 3.57 per 100 procedures (95% confidence interval , 2.12–5.93). Summary event measures were 1.86 per 100 procedures (95% CI, 0.84–4.04) with unilateral lymph node dissection and 3.46 per 100 procedures (95% CI, 1.20–9.56) with bilateral lymph node dissection. In Korea, the trend in the incidence rate of postoperative hypoparathyroidism paralleled that of the incidence rate of thyroid cancer, increasing from 2.6 per 100,000 in 2007 to 7.3 per 100,000 in 2012. Event measures from individual studies were quite variable and, in most instances, were based on very small numbers of events; however, summary measures did not vary much by extent of thyroid or lymph node resection.
A meta-analysis of laryngeal nerve palsy (a cause of unilateral vocal cord paralysis and hoarseness) lasting more than 6 months produced a summary event measure of 1.46 per 100 procedures. Although the individual study measures were less variable than those for hypoparathyroidism, the measures were based on very small numbers of events. Patients who have total thyroidectomy also require lifelong thyroid-replacement therapy and corresponding blood level monitoring. In patients who undergo total thyroidectomy, the process of optimizing hormone replacement therapy and the resultant changes in other medications, weight, or estrogen status may cause iatrogenic hypothyroidism or hyperthyroidism.
Patients with malignant nodules have additional risks if they receive radioactive iodine therapy. Studies of harms of radioactive iodine treatment addressed the risk of second primary malignancy and permanent harms on the salivary glands. The authors concluded, from the available eight studies, that there is a small increase in primary second malignancies, of 12 to 13 excess cancers per 10,000 patients. The authors expressed some concern with that estimate, though, given differences in study designs, reporting of administered doses, and the fact that changes in indication and dose had occurred over time. The most-common permanent salivary adverse effect was xerostomia (dry mouth), which is a risk factor for dental caries; the percentage of affected individuals ranged from 2.3% to 35%. Xerostomia increases the chance of dental decay, demineralization of teeth, tooth sensitivity, and oral infections.
The harms of surgery and radioactive iodine treatments raise concerns because many of these treated cancers may not progress to cause morbidity and mortality. There are no randomized controlled trials of thyroid cancer screening that could be used to estimate overdiagnosis, but it is clear from ecological data that thyroid cancer screening results in detection of thyroid cancers that would not have been diagnosed otherwise. Increases in incidence without changes in mortality in South Korea and other countries in which opportunistic thyroid cancer screening occurs cannot be explained by changes in treatment or risk factor prevalence over the years. Investigators measured thyroid cancer overdiagnosis by studying cancer registry data from high-income countries, estimating age-specific trends in thyroid cancer incidence in the 1960s before ultrasound was introduced, then comparing the shape of the age-specific curves since the 1980s. The investigators reported the rate of overdiagnosis in the United States to be increasing and estimated that it accounted for 77% of thyroid cancer cases. The estimation for overdiagnosis in South Korea was 90% of thyroid cancers cases.
Autopsy studies also lend credence to overdiagnosis resulting from thyroid cancer screening. A 2014 review of 15 autopsy studies reported a 12% yield of papillary thyroid cancers, although the range across studies was wide (1%–36%). Natural history studies have demonstrated the slow-growing nature of thyroid tumors, tumor stability, and low potential for recurrence.