Comparison of Ultrasonography and Amniotic Fluid Cytology in Screening for Suspected Neural Tube Defects in Intrauterine Fetal Demise at Kenyatta National Hospital

Abstract

Background: Neural tube defects are a cause of intrauterine fetal demise or stillbirth globally. Ultrasonography screening for NTDs in IUFDs poses limitations as non-diagnostic sonogram yields have been shown from studies among macerated IUFDs in screening for brain and heart defects. Amniotic fluid cytology is inexpensive, rapid, and complements ultrasonograms on screening for open NTDs and would improve the sensitivity and specificity of screening. The amniocentesis procedure is safe and feasible, carrying low risks or complications for mothers. Studies have been performed on intrauterine fetal demise in various populations, but few in Kenya. There are limited facts on the frequency of neural tube defects in Kenya and studies have recommended more studies and surveillance monitoring systems. Objective: To detect exfoliated neural cells from suspected amniotic fluid smears, to identify positive IHC-stained amniotic fluid cellblocks as confirmatory for open NTDs, and to compare suspected fetal anatomy sonograms for NTDs with their amniotic fluid cytology results. Methodology: This was a cross-sectional descriptive study. Ultrasonograms and amniotic fluid samples were collected from 77 pregnant women with confirmed IUFDs over four months (Dec 2021 – Mar 2022) by purposive sampling method. Clinical and socio-demographic data was collected using a structured questionnaire. Amniotic fluid samples were processed using manual liquid-based cytology, routine staining, and immunohistochemistry staining. Kappa statistical analysis and cross-tabulation were used to check the level of agreement and significance between the diagnostic methods respectively. Results were presented as tables, figures, bar graphs, and pie charts. Results: A total of 77 gravid mothers with IUFDs were recruited and out of these, 26 had a complete dataset. Multiple imputation analysis was done to replace the missing data with substitute values to retain information about the dataset. Out of the 77 ultrasonograms and amniotic fluid samples, 20 ultrasonographic scans showed abnormal fetal anatomy of the IUFDs and 29 amniotic fluid samples were positive for neural tube defects respectively. The prevalence of neural tube defects among IUFDs was 26%. There was a significant difference between ultrasonography and amniotic fluid cytology (p = 0.003), and a comparability of 0.82 kappa statistic. Conclusion: The study provides essential information comparison between ultrasonography with amniotic fluid cytology for screening pregnant women with IUFDs. The data showed a significant difference between ultrasonography and amniotic fluid cytology screening for NTDs, with cytology performing better. Amniotic fluid cytology in this study complements ultrasonography in the detection of neural tube defects in IUFDs