Clinical Implications of Different Thyroid-Stimulating Hormone (TSH) Reference Intervals between TSH Kits for the Management of Subclinical Hypothyroidism
Article information
In 2023, the Korean Thyroid Association (KTA) released guidelines for the management of subclinical hypothyroidism [1]. The guidelines advise treatment for patients under 70 years old with thyroid-stimulating hormone (TSH) levels >10 mIU/L, while suggesting observation for those aged 70 and above [2]. Notably, the guidelines introduce a new upper limit for normal TSH levels of 6.8 mIU/L among Koreans. Previous guidelines have recommended that TSH reference ranges should be established based on regional research results, acknowledging that TSH standards might vary by region [3,4]. Accordingly, drawing upon data from the 2016 Korea National Health and Nutrition Examination Survey, the KTA defined the TSH reference range for Koreans as 0.6 to 6.8 mIU/L [5,6].
However, TSH measurements can vary significantly depending on the assay kit used. The KTA guidelines are based on a study using a single kit (Roche), but recognize the limitations in addressing inter-kit differences accurately [1,2]. To overcome this limitation, the KTA recently conducted a study that directly compared various TSH kits, including both chemiluminescence immunoassay (CLIA) and radioimmunoassay (RIA) methods, which are widely used in Korea [7]. Methodologically, this study was similar to projects such as the Canadian Laboratory Initiative on Pediatric Reference Intervals (CALIPER) study, which compared reference ranges among different kits [8].
The study analyzed 100 residual blood samples, which had TSH levels ranging from 0 to 20 mIU/L, using eight different kits, including Roche(Basel, Switzerland), Abbott (Abbott Park, IL, USA), Beckman-CLIA (Brea, CA, USA), Siemens (Munich, Germany), Beckman-RIA (Brea), BRAHMS (Berlin, Germany), RIAKEY (Seoul, Korea), and Izotop (Budapest, Hungary) [7]. Correlation and regression analyses were performed on repeated measurements, providing slopes, intercepts, and 95% confidence intervals for the regression equations (Table 1).
Using Roche as the reference kit, the results for most kits demonstrated excellent correlations (R>0.99). However, the slopes varied between 0.75 and 1.06, indicating differences depending on the kit type. Consequently, the estimated values for the medical decision point of TSH corresponding to a level of 6.8 mIU/L measured using the Roche kit ranged between 5.2 and 7.4 mIU/L. For a level of 10.0 mIU/L using the Roche kit, the values for other kits ranged between 7.7 and 10.8 mIU/L, demonstrating considerable variation in the estimates across kits [7]. In general, the kit with the lowest TSH measurements was Abbott, while the kit with the highest measurements was Beckman-RIA. Previous studies, primarily comparing Roche and Abbott, consistently reported that Roche’s TSH measurements tended to be higher than Abbott’s, and this should be kept in mind [9,10].
In conclusion, it is of paramount importance for clinicians to maintain awareness of the TSH kit they employ and its characteristics compared to the Roche kit used in guideline development. Considering the pivotal role of TSH as a primary indicator in initiating and adjusting thyroid hormone therapy, widespread dissemination of this information is essential. This will help clinicians make well-informed decisions in the context of real-world clinical scenarios.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.