Evaluation of amniotic fluid cell filtration: an experimental approach to early amniocentesis

Prior to a prospective application of amniotic fluid (AF) cell filtration to early amniocentesis, we tested the technique on a surplus from mid-trimester samples. By using the same sample size of 5 ml in experiments with a filter and in routine diagnostic procedures (control), we evaluated an optimal filter system. The prolonged culture time of filtered cells and the reduced number of clones are most probably due to mechanical stress (filtration pressure), whereas loss of the cells by adhesion to the filter system, and an AF-free culture medium (growth factors) are suggested to be less important. The AF cells are very sensitive to mechanical stress. Slow filtration (< or = ml AF/min) through filters with a high porosity and the largest possible pore size should be preferred. A mixed cellulose ester filter membrane with a pore size of 5.0 microns proved to be the most efficient, allowing harvest of the filtered cells after only a slight prolongation of the culture time (+2.4 days) compared with unfiltered aliquots. A filter set with a bypass connected by three-way taps allows cell filtration during either aspiration or reinjection of the AF. Cell filtration after amniocentesis and consecutive reverse flushing of the membrane with the appropriate amount of culture medium proved to be the best with regard to easy handling and reducing the risk of bacterial contamination.