Several systematic reviews and meta-analyses have investigated the effect of levothyroxine (LT4) therapy in pregnant women with subclinical hypothyroidism (SCH). However, all these studies have clinical or methodological problems (such as adopting the old 2011 American Thyroid Association [ATA] diagnostic criteria, directly combining randomized controlled trials [RCTs] and cohort studies for meta-analysis, and so on), and cannot provide accurate and satisfactory results. Thus, we performed this updated systematic review, meta-analysis and trial sequential analysis (TSA) to assess the effect of LT4 therapy in pregnant women with SCH, with the goal of providing more accurate and reliable evidence for clinical practice.
We searched nine databases from inception to February 2022. The search strategy targeted the RCTs and cohort studies on pregnancy, neonatal and childhood outcomes following LT4 treatment in pregnant women with SCH based on the new 2017 ATA diagnostic criteria. We performed meta-analyses of RCTs and cohort studies separately, and further performed meta-analyses by excluding studies with high risk of bias. TSA was performed to test whether the current evidence was sufficient, and the quality of evidence was evaluated using the GRADE method.
A total of 9 RCTs and 13 cohort studies comprising 11273 pregnant women with SCH were included. There were no statistically significant differences between LT4 group and control group in all primary and secondary outcomes, such as preterm delivery (RR=0.46, 95%CI: 0.19-1.09, P=0.08, I 2 = 65%), miscarriage (RR=0.36, 95%CI: 0.13-1.03, P=0.06, I 2 = 38%), gestational hypertension (RR=0.91, 95%CI: 0.58-1.43, P=0.69, I 2 = 0%), preeclampsia (RR=1.10, 95%CI: 0.61-1.97, P=0.76, I 2 = 0%), gestational diabetes (RR=0.80, 95%CI: 0.51-1.25, P=0.32, I 2 = 34%), and so on. TSA showed that the results for all outcomes were insufficient and inconclusive. According to GRADE, the evidences for four outcomes (miscarriage, gestational hypertension, gestational diabetes, and small for gestational age) were rated as moderate quality, while the evidences for the other outcomes were rated as low or very low quality.
Unlike previous systematic reviews and meta-analyses, our study found no evidence of benefit of LT4 therapy on pregnancy, neonatal and childhood outcomes in pregnant women with SCH.
PROSPERO, https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42022321937, identifier CRD42022321937.