Stillbirths at Term: Case Control Study of Risk Factors, Growth Status and Placental Histology

Abstract

ObjectiveTo investigate the proportion of stillbirths at term associated with abnormal growth using customized birth weight percentiles and to compare histological placental findings both in underweight stillborn fetuses and in live births.MethodsA retrospective case-control study of 150 singleton term stillbirths. The livebirth control groups included 586 cases of low-risk pregnancies and 153 late fetal growth restriction fetuses. Stillbirths and livebirths from low-risk pregnancies were classified using customized standards for fetal weight at birth, as adequate for gestational age (AGA; 10-90th percentile), small (SGA; <10th percentile) or large for gestational age (LGA; >90th percentile). Placental characteristics in stillbirth were compared with those from livebirths using four categories: inflammation, disruptive, obstructive and adaptive lesions.ResultsThere was a higher rate of SGA (26% vs 6%, p<0.001) and LGA fetuses (10.6% vs 5.6%, p<0.05) in the stillbirth group. Among stillbirth fetuses, almost half of the SGA were very low birthweight (≤3°percentile) (12% vs 0.3%, p<0.001). The disruptive (7.3% vs 0.17%;p<0.001), obstructive (54.6% vs 7.5%;p<0.001) and adaptive (46.6% vs 35.8%;p<0.001) findings were significantly more common in than in livebirth-low risk. Placental characteristics of AGA and SGA stillbirth were compared with those of AGA and FGR livebirth. In stillbirths-SGA we found a higher number of disruptive (12.8% vs 0%; p<0.001), obstructive (58.9% vs 23.5%;p<0.001) and adaptive lesions (56.4% vs 49%; p 0.47) than in livebirth-FGR.ConclusionThe assessment of fetal weight with customized curves can identify fetuses which have not reached their genetically determined growth potential and are therefore at risk for adverse outcomes. Placental evaluation in stillbirths can reveal chronic histological signs that might be useful to clinical assessment, especially in underweight fetuses.

Human leukocyte antigen DQ2/DQ8 positivity in women with history of stillbirth

“…A number of studies showed that in women with CD recurrent pregnancy loss (RPL), fetal intrauterine deaths, fetal growth restriction (FGR), and small for gestational age (SGA) fetuses are more frequent. 13 A recent meta-analysis of 10 cohort studies shows that women with celiac disease (both treated and untreated) have a significantly higher risk of stillbirth (OR: 4.84). Beyond the nutrient deficiency characterized by lack of absorption of elements such as zinc, selenium, and folic acid, the role of anti-transglutaminase antibodies to impair trophoblast invasiveness and endometrial endothelial cells differentiation has been suggested as destructive element of early placentation.…”

“…10,11 We have previously observed that fetuses which have not reached their genetically determined growth potential are at risk for adverse outcomes 12 and that some placental abnormalities (particularly obstructive lesions) may play a crucial role in underweight stillborn fetuses at term. 13 A recent meta-analysis of 10 cohort studies shows that women with celiac disease (both treated and untreated) have a significantly higher risk of stillbirth (OR: 4.84). 9 A 3.6 times higher prevalence of HLA-DQ2/DQ8 positivity was observed in women with RPL compared to women with at least two previous uncomplicated term pregnancies.…”

Bayesian point process modeling to quantify excess risk in spatial epidemiology

“…To date, studies assessing risk factors associated with stillbirth have largely been population-based case-control studies (Flenady et al, 2011 provide a recent systematic review; recent studies include Mecacci et al, 2016 andKumbhare andMaitra, 2016), with a few studies only incorporating minimal geospatial information into the analysis (for example, DeFranco et al, 2015 andHall et al, 2014). For example, DeFranco et al, 2015 assessed whether exposure to increased levels of fine particulate air pollution (P M 2.5 ) was associated with stillbirth; geographic coordinates of maternal residence for each birth were linked to the nearest air monitoring station.…”

“…To date, studies assessing risk factors associated with stillbirth have largely been population-based casecontrol studies (Flenady et al, 2011 provide a systematic review; recent studies include Kumbhare andMaitra, 2016 andMecacci et al, 2016), with a few studies incorporating minimal geospatial information into the analysis (for example, Hall et al, 2014 andDeFranco et al, 2015). To our knowledge, no analysis of point patterns associated with stillbirth exist.…”

“…To date, studies assessing risk factors associated with stillbirth have largely been population-based case-control studies (Flenady et al, 2011 provide a recent systematic review; recent studies include Mecacci et al, 2016 andKumbhare andMaitra, 2016), with few studies incorporating geospatial information, although minimal information, into the analysis (for example, Hall et al, 2014 andDeFranco et al, 2015). Logistic regression was the corresponding analytical method used to identify risk factors; the outcome events (stillbirth and live birth), which occurred at observed spatial locations, were modeled as conditionally independent binomial outcomes and the spatial variation inherent in the locations associated with the events within the study specific region was ignored.…”