Does the use of letrozole in infertility treatment increase the risk of foetal malformations? A systematic review and meta-analysis of randomized controlled trials.

INTRODUCTION

Despite recent studies, the use of letrozole (LTZ) in reproductive medicine is still limited due to previous concerns regarding foetal safety. Purpose of this systematic review and meta-analysis was to assess whether the use of LTZ prior to assisted reproduction may increase the rate of foetal malformations.

METHODS

MEDLINE, Scopus, ClinicalTrials.gov, Scielo, EMBASE, the Cochrane Library at the CENTRAL Register of Controlled Trials, LILACS, PROSPERO, conference proceeding, grey literature and international controlled trials registries were searched from inception to July 2024 with no geographical or language restrictions. Randomized controlled trials (RCTs) seeking for congenital malformations and pregnancy losses between LTZ and clomiphene citrate (CC), gonadotropins, natural conception (NC) or other agents were included. Random-effect meta-analysis was performed adhering to PRISMA statement. Cochrane criteria were employed for quality of evidence and risk of bias assessment and GRADE criteria for certainty of evidence. Primary outcome was the incidence of congenital malformations. Secondary outcome was the pregnancy loss rate (PLR).

RESULTS

Twenty studies (6679 patients) were included. We did not find an increase in congenital malformations with LTZ vs. CC (risk ratio (RR) 1.69 [95% CI 0.51 to 5.58]; very low certainty). Subdividing according to the type of conception, no increased risk was retrievable among treatment arms for timed intercourse (RR 3.61 [95% CI 0.72 to 18.14]; low certainty), intrauterine insemination (IUI) (RD 0.00 [95% CI -0.01 to 0.01]) and no registered events for frozen embryo transfer (FET) (0/50 vs 0/50). Moreover, no significant difference with NC (RR 1.53 [95% CI 0.15 to 15.89]; very low certainty), gonadotropins (RR 0.67 [95% CI 0.11 to 3.97]; very low certainty), or berberin (RR 1.77 [95% CI 0.07 to 42.63]; very low certainty) were retrieved. Regarding the PLR, no significant differences between LTZ and CC (RR 1.17 [95% CI 0.91 to 1.50]; low certainty), LTZ and NC (RR 0.88 [95% CI 0.28 to 2.75]; very low certainty) were retrieved. Compared to gonadotropins and berberin, a slightly lower RR for LTZ was noted (RR 0.61 [95% CI 0.40 to 0.93]; very low certainty, and RR 0.76 [95% CI 0.40 to 1.42]; very low certainty respectively).

CONCLUSIONS

Altough current evidence does not suggest an increased risk of congenital malformations or pregnancy loss associated with the use of LTZ for timed intercourse, IUI, or FET, the certainty of evidence remains very low. Therefore, while available data can be considered somewhat reassuring, uncertainty remains about the true effect of LTZ on these outcomes, stressing the need for further RCTs.