The predictive ability of clinical palpation for estimating amniotic fluid volume in suspected prolonged pregnancy
Buchmann, Eckhart Johannes
Background and objectives In low resource settings, ultrasound scans may not be available for amniotic fluid volume (AFV) assessment as part of fetal evaluation for suspected prolonged pregnancies (≥41 weeks’ gestation). The objectives of this study were: 1) to describe AFV measurements using ultrasound in women with suspected prolonged pregnancies, and to relate these measurements to maternal and fetal factors; 2) to evaluate different clinical palpation methods for estimating AFV; and 3) to determine the ability of clinical palpation to estimate AFV and predict oligohydramnios, using ultrasound-based amniotic fluid index (AFI) as a gold standard, accounting for the influence of maternal and fetal factors. Methods The study included women referred to Chris Hani Baragwanath Academic Hospital from midwife-run antenatal clinics because of concern about prolonged pregnancy (gestational age ≥41 weeks). On arrival at hospital, the women had real-time ultrasound assessment of AFI by an experienced ultrasonographer. The researcher, blinded to the AFI result, estimated AFV by abdominal palpation using ballottability of fetal parts, uterine fluctuance, uterine irritability, easily felt fetal parts, and a general impression of AFV. After recording the palpation findings, the researcher made a best estimate of gestational age for each woman based on the last menstrual period, early pregnancy ultrasound scans if available, or other relevant clinical information. Oligohydramnios was defined as an AFI <5 cm. Results One hundred women participated, of whom 45 had a best estimate gestational age ≥41 weeks. The mean AFI was 8.1±4.3 cm; 23 women had an AFI <5 cm. Twenty women were HIV infected. In univariable and multivariable linear regression analysis, HIV infection and gestational age were inversely associated with AFI at a P value <0.05. On abdominal palpation, the symphysis-fundal height, uterine fundal fetal parts ballottement, and presenting part ballottement were significantly positively associated with AFI on univariable and multivariable linear regression analysis. For the binary outcome of oligohydramnios, only presenting part ballottement was associated with an AFI <5 cm (negatively), both crudely, and adjusted for gestational age and HIV infection using logistic regression analysis. For women with a fetal head fully palpable (‘five fifths’) above the pubic symphysis (n=55), an inability to ballot the presenting part had a sensitivity of 73%, specificity of 64%, and a negative predictive value of 90% for an AFI <5 cm. Conclusion While fetal part ballottement and symphysis-fundal height measurement showed significant associations with AFI, the predictive value of clinical palpation for oligohydramnios was poor. However, in settings where real-time ultrasound technology is unavailable, the assessment of presenting part ballottement may be of value in women with suspected prolonged pregnancy. In clinical settings similar to those in this study, a ballottable fetal head gives 90% assurance of normal AFV.