Reply to Reconsidering the Impacts of Steatotic Liver Disease on Adverse Pregnancy Outcomes

We sincerely appreciate the interest of Tang and colleagues in our study on the association between trimester-specific steatotic liver disease and adverse pregnancy outcomes. We are delighted to have the opportunity to further discuss those thoughtful questions they have raised [1]. Regarding the diagnostic accuracy of de novo steatosis in the third trimester [2], we consulted the ultrasound physicians in our hospital. While it is true that gestational weight gain and venous congestion could theoretically influence imaging interpretation, several factors support the reliability of our assessment. First, the liver is a relatively large abdominal organ, allowing evaluation through multiple ultrasound planes. In practice, fetal position and uterine enlargement seldom impede adequate imaging of the liver during the third trimester. Second, to minimise overdiagnosis, our ultrasound physicians are highly experienced in recognising physiological changes in pregnancy, including increased subcutaneous fat. The appropriate probes would be selected and multiple imaging indicators would be integrated, such as liver–kidney echo contrast, vascular clarity and deep attenuation, to enhance diagnostic accuracy. Last, our data show that de novo steatosis in the third trimester correlates consistently with other clinical indicators: most of these women (67.78%) had excessive gestational weight gain, and their serum triglyceride levels rose from a median of 1.15–3.05 mmol/L, an increase second only to that observed in the persistent steatosis group. This parallel rise in hepatic and systemic lipid levels supports the biological plausibility of our imaging findings. As to the remaining two questions, particularly, whether de novo steatotic liver disease contributes to postpartum haemorrhage, we offer the following perspective. First, pregnancy per se is a hypercoagulable situation [3]. Current evidence does not strongly suggest that simple steatosis significantly exacerbates this condition; one study in non-pregnant adults found comparable haemostatic profiles between patients with non-cirrhotic NAFLD and controls, indicating a minimal effect of steatosis on coagulation [4]. However, we did not evaluate the liver stiffness via VCTE technology; the severity of the liver fibrosis was not so clear. Although our study shows a statistical association between de novo steatosis in the third trimester and postpartum haemorrhage, we suspect this link may be mediated by pregnancy-associated hypertension (PAH) [2]. This is supported by the association analyses between PAH and the de novo steatosis in the third trimester. The PAH that occurred in the second trimester was significantly associated with de novo steatosis in the third trimester [2], a pattern consistent with findings from prior studies [5]. Given that PAH is a well-established risk factor for postpartum haemorrhage [6], it may represent a key intermediary in the observed association. In addition, only one large-scale study to date, involving over 18 million women, has reported that NAFLD is associated with higher odds of postpartum haemorrhage, even after excluding cases with pre-eclampsia, eclampsia and/or HELLP syndrome (haemolysis, elevated liver enzymes and low plates) [7]. In both that study and ours, the mean maternal age was around or even beyond 30 years. We speculate that the advanced maternal age and the increased prevalence of MASLD may coexist in the analysis. Although we adjusted for maternal age, the independent effect of age on postpartum haemorrhage remains difficult to fully disentangle [8]. With a larger sample size in the future, stratification by age could help clarify the association between MASLD and postpartum haemorrhage. The authors declare no conflicts of interest. This article is linked to Tang papers. To view this articles, visit https://doi.org/10.1111/liv.70499. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.