Serum thyrotropin determinations: comparison of second- and third-generation assays.

Objective: To compare thyrotropin (thyroid-stimulating hormone or TSH) levels determined by second- and third-generation assays in sera extracted from blood samples withdrawn randomly from 34 subjects attending outpatient clinics.

Methods: Serum specimens were separated into two fractions and were frozen and stored at -20 degrees C for subsequent determination of TSH concentrations by second-and third-generation assays. The results were analyzed statistically.

Results: Of the 34 serum samples, 23 TSH levels were within normal ranges on both assays, whereas 9 samples showed subnormal TSH concentrations with both second- and third-generation assays. The other two samples revealed mildly increased TSH levels by the third-generation assay but high normal levels with the second-generation assay. Moreover, mean TSH values determined with second- and third-generation assays for all 34 serum samples were not significantly different (1.48 versus 1.66 mU/L; P>0.05). Finally, a highly significant correlation (r = 0.987) was also noted between the two assays on simple linear regression.

Conclusion: Thus, serum TSH determinations by both the second- and third-generation assays are almost identical, and attempting to achieve greater sensitivity to detect lower levels by further modifications in third-, fourth-, and even later generation assays may be an expensive and fruitless endeavor because of the inability of these methods to discern the thyroid functional state any better than the second-generation assay.