The effect of different levothyroxine administration regimens on thyroid hormone levels: a systematic review, pairwise, and network meta-analysis

Abstract

ABSTRACT Introduction This systematic review aimed to compare the effect of alternative levothyroxine administration regimens on thyroid hormone levels and patient-reported outcomes (PROs) among adults with hypothyroidism. Methods We searched PubMed, Embase, CENTRAL, CINAHL, LILACS, SciELO, Scopus, Web of Science, OpenGrey, ProQuest, ClinicalTrials.gov, and ICTRP from inception to May/2023 for randomized controlled trials (RCTs). We assessed the risk of bias with Cochrane Risk of Bias 2.0 tool. We analyzed TSH levels by pairwise and network meta‐analyses (NMA). The FT4 levels and PROs were qualitatively assessed. Results We included 14 RCTs (906 participants) comparing different regimens, as bedtime vs. before breakfast. A total of 12 RCTs were at high risk of bias. Seven RCTs were included in the TSH meta-analysis, where the mean difference (MD) and 95% confidence interval (CI) were as follows: bedtime vs before breakfast (4 RCTs) 0.69 (−1.67–3.04), I2 = 92%, very low certainty evidence; weekly dose vs before breakfast (2 RCTs) 1.68 (0.94–2.41), I2 = 0%, low certainty evidence; and at breakfast vs before breakfast (1 RCT) 0.65 (−1.11–2.41), very low certainty evidence. The NMA showed no evidence of differences in TSH level with different regimens. Conclusion The evidence is insufficient to determine the most effective levothyroxine administration regimen for hypothyroidism. Systematic review registration PROSPERO – CRD42021279375