INTERMACHINE ERROR IN FETAL DOPPLER FLOW VELOCITY WAVE-FORM ANALYSIS - A PRELIMINARY-REPORT

Objective: The aim of the study was to establish inter-machine reprodu cibility of the pulsatility index and resistance index calculated from Doppler flow velocity waveforms from the fetal umbilical artery. Meth ods: Studies of fetal umbilical artery waveforms from two patients at 36 weeks' gestation were performed by a single operator. A Diasonics D RF 400 machine was paired with one of two Toshiba 270 machines and stu dies were alternated between machines during a session. Between 8 and 15 examinations were performed per machine per session. The mean of fi ve consecutive waveforms was measured on line for each examination. Re sults: The pulsatility index from the same patient was significantly d ifferent between machines (session 1: 1.14 vs 0.78, P = 0.0013; sessio n 2; 1.08 vs 0.72, P = 0.0009; session 3: 1.17 vs 1.00, P = .0001; ses sion 4: 1.20 vs 0.86, P = 0.0003) but was not significantly different using either the Diasonics machine (1.14 vs 1.08 vs 1.17, P = 0.5949) or a Toshiba machine with different probes (0.78 vs 0.72, P = 0.4598) on the same patient at different sessions. The resistance index from t he same patient did not differ between machines (0.66 vs 0.62, P = 0.2 778; 0.64 vs 0.66, P = 0.3367). Conclusions: There may be large interm achine variations in the on-line calculation of the pulsatility index from the fetal umbilical artery. Caution is warranted when comparing t he pulsatility index either between centers or from the same center wh en different equipment is used. Further studies are necessary to confi rm that the resistance index is reproducible with different machines.