Endocrine Abstracts

Background: Variability in oestradiol profiles across hormone replacement therapy (HRT) preparations has clinical implications. However, short-term variability across preparations has not been examined previously. Herein, we assessed short-term reproductive hormone variability in postmenopausal women on different HRT preparations.

Methods: Reproductive hormone levels were measured in 35 postmenopausal women using gel, oral, patch, or spray HRT. For gel, oral, and spray users, serial venous sampling was performed every 15mins over 90mins. Variability was quantified using peak–trough changes and coefficient-of-variation (CV%), with the oestradiol assay CV ≤7.7%. Hormonal differences were assessed using Kruskal–Wallis test, and associations with Spearman’s r.

Results: Age [mean (SD)] was 56.7 (4.1) years and BMI 24.0 (2.7) kg/m². Oestradiol-equivalent doses ranged from low–medium for oral (n = 4) and spray (n = 3), low–high for gel (n = 15), and medium–high for patch (n = 15). Baseline oestradiol was negatively correlated with FSH for gel (r = -0.56, P = 0.03) and patch (r = -0.57, P = 0.04), with weaker correlations for LH (gel: r = -0.54, P = 0.06; patch: r = -0.50, P = 0.06). Short-term oestradiol variability differed by preparation: peak–trough changes [median (IQR)] were 144.8pmol/l (80.7–324.6) for gel, 177.9pmol/l (115.8–280.1) for oral, and 395.9pmol/l (130.2–891.6) for spray. Spray also exhibited the highest oestradiol CV% [54.1% (36.7–78.3)], compared with oral [17.1% (10.7–24.2), P = 0.04] and gel [15.1% (2.8–17.0), P < 0.01] preparations. Oral and gel did not differ significantly (P = 0.08).

Discussion: In postmenopausal women, oestradiol negatively correlated with FSH and LH, consistent with negative feedback. Short-term oestradiol variability differed by HRT preparation. Spray exhibited the greatest fluctuations, followed by gel and oral preparations. Variability exceeded intra-assay CV% confirming pharmacokinetic rather than analytical effects. Our data shows that single time-point measurements may not represent prevailing oestradiol levels and may not be suitable for monitoring or adjusting HRT, which should instead be guided by symptoms.