OBJECTIVES: Minimally invasive thyroidectomy (MIT), both laparoscopically and robotically, is increasingly being implemented for cancer care in the United States. Rates of conversion to an open procedure and risk factors for conversion at a national level are unknown.

METHODS: Patients with thyroid cancer who underwent MIT between 2010 to 2019 were selected from the National Cancer Database, and conversion to open cases was identified. Average annual volume of thyroidectomies (TV), including both open and MIT, and cumulative volume of MIT for each year (MIT-CV) was calculated for all facilities A facility-clustered logistic regression model was used to determine patient, tumor, procedure, and facility level factors independently associated with conversion.

RESULTS: A total of 7399 MIT cases were identified, 83.4% of which were laparoscopic. Conversion rates were initially higher with the laparoscopic approach and decreased overtime (from 13.3% in 2010 to 4.8% in 2019, p<0.001), while they remained stable with the robotic approach (from 4.6% in 2010 to 6.1% in 2019, p=0.379). Although converted cases had larger lesions (16mm[IQ:7-30];vs13mm[IQ:6-24];p=0.037), tumor size, surgery extension – total thyroidectomy and nodal dissection were not associated with conversion. Most conversions (54.1%) occurred in facilities with a MIT-CV <6 cases, which corresponded to the lowest quartile. On multivariable analysis, the risk of conversion decreased as the MIT-CV increased (OR:0.49;95%CI[0.31-0.77]) for MIT-CV 7 to 19 cases; (OR:0.17;95%CI[0.09-0.34]) for MIT-CV 20 to 51 cases and (OR:0.20;95%CI[0.07-0.56]) for MIT-CV >51 cases. When adjusted for MIT-CV, neither year or TV were associated with conversion.

CONCLUSIONS: As MIT for thyroid cancer is being implemented nationwide, conversion rates are decreasing. Conversion rates decreased as the facility experience with MIT increased, denoting a facility learning curve, while TV was not associated with conversion. Other adverse events such as laryngeal nerve injury need to be evaluated to determine the impact of MIT learning curve.