Derivation of a multivariate reference range for pituitary thyrotropin and thyroid hormones: diagnostic efficiency...

Area of Science:

• Endocrinology
• Clinical Chemistry
• Biostatistics

Background:

• Thyroid hormone and TSH measurements are conventionally interpreted separately.
• Subclinical thyroid dysfunction classification relies solely on abnormal TSH levels.
• This study investigates a composite multivariate approach for disease classification.

Purpose of the Study:

• To examine a composite multivariate approach for classifying thyroid dysfunction.
• To compare the diagnostic efficiency of multivariate reference limits with conventional methods.

Main Methods:

• Derived bivariate and trivariate reference limits from a thyroid-healthy control group (n=271).
• Applied these limits to a diverse clinical sample (n=820) from a prospective study.
• Compared the diagnostic efficiency of multivariate methods against conventional separate parameter interpretation.

Main Results:

• Composite reference limits (bivariate and trivariate) reclassified 6.3% and 12% of thyroid dysfunctions, respectively.
• The bivariate method reclassified 26% of established dysfunctions to euthyroid status.
• Multivariate approaches revealed discrepancies compared to separate univariate reference intervals.

Conclusions:

• Statistical analytic techniques significantly influence thyroid disease classification.
• Divergencies challenge the use of TSH as both a sensitive screening and reliable classification tool.
• A composite multivariate approach may offer a more comprehensive assessment of thyroid function.