OBJECTIVE: Our purpose was to study the effect of operator transducer pressure on amniotic fluid index and single deepest pool measurements and their intraobserver and interobserver variability. STUDY DESIGN: Forty subjects in the third trimester with intact membranes were studied by two observers at three predetermined operator pressures. Pressure was measured by interfacing a flexible water-filled reservoir connected to a manometer between the maternal abdomen and the ultrasonography transducer. The amniotic fluid index and single deepest pool were measured on screen by another operator, and data were analyzed by calculating components of variance. RESULTS: Compared with medium pressure, low pressure resulted in a 13% increase in amniotic fluid index (p < 0.001), and high pressure resulted in a 21% fall in amniotic fluid index (p < 0.001). The single deepest pool was less sensitive to pressure, with a 11% increase at low pressure (p < 0.001) and a 16% fall at high pressure (p < 0.001). Intraobserver error was 17% for amniotic fluid index and 18% for single deepest pool when operator pressure was controlled at medium pressure; this increased to 28% and 24%, respectively, when pressure was not controlled (p < 0.01). Interobserver error was consistently very low. CONCLUSION: Both techniques are highly sensitive to the pressure applied to the maternal abdomen during scanning. The reproducibility of amniotic fluid index and single deepest pool is comparable. The use of the same observer for serial measurements of amniotic fluid index would appear to be less important than careful attention to transducer pressure.
OBJECTIVE: Our purpose was to study the effect of operator transducer pressure on amniotic fluid index and single deepest pool measurements and their intraobserver and interobserver variability. STUDY DESIGN: Forty subjects in the third trimester with intact membranes were studied by two observers at three predetermined operator pressures. Pressure was measured by interfacing a flexible water-filled reservoir connected to a manometer between the maternal abdomen and the ultrasonography transducer. The amniotic fluid index and single deepest pool were measured on screen by another operator, and data were analyzed by calculating components of variance. RESULTS: Compared with medium pressure, low pressure resulted in a 13% increase in amniotic fluid index (p < 0.001), and high pressure resulted in a 21% fall in amniotic fluid index (p < 0.001). The single deepest pool was less sensitive to pressure, with a 11% increase at low pressure (p < 0.001) and a 16% fall at high pressure (p < 0.001). Intraobserver error was 17% for amniotic fluid index and 18% for single deepest pool when operator pressure was controlled at medium pressure; this increased to 28% and 24%, respectively, when pressure was not controlled (p < 0.01). Interobserver error was consistently very low. CONCLUSION: Both techniques are highly sensitive to the pressure applied to the maternal abdomen during scanning. The reproducibility of amniotic fluid index and single deepest pool is comparable. The use of the same observer for serial measurements of amniotic fluid index would appear to be less important than careful attention to transducer pressure.