Figure 1 – Summary of the methodology
All pregnant women with a period of gestation of less than 12 weeks who registered with public health midwives from early July-end of September 2019 were recruited from all MOH areas in Anuradhapura district for the study. Collection of socio-demographic data was performed by medical undergraduates in the third year of training, and clinical data collection and the examination were performed by MBBS-qualified medical officers. Baseline data was collected by using an interviewer-administered questionnaire. Pre-tested interviewer administered questionnaire was used for data collection on sociodemographic, obstetric, gynecological and medical backgrounds. Those with a history of diabetes mellitus, uncertain period of amenorrhea (POA), known liver diseases (except NAFLD) and history of using steatogenic drugs were excluded from the initial recruitment.
Clinical examination was performed to elicit any signs of pre-existing diseases. Blood pressure was recorded using a high precision automated blood pressure measuring instrument (Omron Corporation) as the mean of two readings taken 5 min apart from both arms it and was categorized as normal and high values for the first and second trimesters according to the NICE guidelines 2019 28. Height (Ht), weight (Wt), body mass index (BMI), waist circumference (WC), hip circumference (HC) and waist to hip ratio (WHR) was taken as anthropometric measurements.
Weight was measured by using a digital weighing scale and height was measured by using a portable stadiometer. According to standard formula of BMI, categorized as overweight, obesity and morbid obesity and according to standard cutoff levels for Asian population by WHO guidelines. Waist circumference (WC) was measured by placing a non-stretchable fibre-glass measuring tape around the waist midway between the last rib and iliac crest with the subject in the standing position. Two measurements were taken for improvement of the accuracy of measurement. Hip circumference (HC) was measured the maximum circumference of buttocks. Two measurements were taken and sum of the measurement will be taken as WC and HC. Some data was obtained from maternal pregnancy record. This entered data was standardized to make sure the routine data are high quality 29. During first trimester of pregnancy there are very minimal changes of weight gain, and changes of waist circumference. Therefore we use standard calculation methods and ranges for Asian adult population to calculate BMI, cut off levels of obesity, waist circumference pregnancy period.
Baseline investigations were done to identify common metabolic complications and to exclude any pathological conditions not related to NAFLD. These investigations include routine basic investigations of mothers should be done their antenatal period and investigations for diagnose dyslipidemia and liver related disorders. Blood samples were collected by using standard guidelines by well qualified public health nursing sisters. Pre requisites for sample collection were informed to mothers prior to the blood collection date. Venipuncture was done at ante-cubital fossa under aseptic conditions and universal precautions. Internal quality control was performed before each and every analytical run. Peer group comparison was done every month during sample collection and period of analysis of collected blood samples.
An oral glucose tolerance test (OGTT) was performed on all participants at the recruitment and second trimester. Diagnoses of diabetes mellitus in pregnancy (DIP) and gestational diabetes mellitus (GDM) were performed using WHO (2016) criteria 30. GDM was defined as fasting blood sugar (FBS) of 92-125 mg/dl and/or 2nd-hour plasma glucose of 153-199 mg/dl by a 75-g glucose test anytime in pregnancy. Those with FBS ≥ 126 mg/dl and/or 2nd-hour plasma glucose ≥ 200 mg/dl were labelled as having diabetes mellitus in pregnancy 30. Level of serum aminotransferaces, gamma glutamyl transferaces and serum lipid level was done as baseline screening tests to identify any liver related pathological conditions and pre-existing dyslipidemia.
After obtaining written informed consent, all eligible study participants were offered an ultrasound scan (USS) abdomen. USS was performed by competent and qualified investigators. Liver echogenicity was compared with the ipsilateral renal cortex and the spleen, and the attenuation of waves, loss of demarcation of the diaphragm and poor demarcation of the intrahepatic architecture were examined. Thus, fatty liver was graded as 20; Grade 0: Normal liver echogenicity, Grade I:  Diffusely increased hepatic echogenicity but periportal and diaphragmatic echogenicity is still appreciable, Grade II:  Diffusely increased hepatic echogenicity obscuring periportal echogenicity but diaphragmatic echogenicity is still appreciable, and Grade III:  Diffusely increased hepatic echogenicity obscuring periportal and diaphragmatic echogenicity.
Although comparison of fatty liver between pregnancy and the general population is not a major objective of the study, a small sample of an age-matched randomly selected females was subjected to USS to overcome the possible bias of early pregnancy liver changes.