Single pre-delivery symphysis-fundal height measurement as a predictor of birthweight and multiple pregnancy

Objective To compare the value of different pre-delivery maternal indices for predicting birth­ weight, and to examine the usefulness of a single pre-delivery symphysis-fundal height measurement for the detection of low birthweight and twin pregnancy or macrosomia. Design Symphysis-fundal height measurements were gathered from 1509 women who had both a singleton delivery and available data of pre-delivery weight, height and mid-upper arm circumference, and from 73 women who had a twin delivery. Setting A district hospital in rural Tanzania. Results Symphysis-fundal height, pre-delivery weight and mid-upper arm circumference, re­ spectively, explain 41 %, 13% and 4% of the observed variation in birthweight. At a cut-off level of 30 cm for symphysis-fundal height, the detection rate for birthweight below 2500 g and 2000 g was 66% and 68% , respectively, and the false positive rate was 9% and 14%, respectively. At a cut-off level of 38 cm for symphysis fundal height the detection rate for twin pregnancy or birthweight > 4000 g was 76% , and the false positive rate was 4% . Conclusion Symphysis-fundal height was a better predictor of birthweight than maternal height, pre-delivery weight or mid-upper arm circumference. It seems justified to investigate the value of a simple tricoloured symphysis-fundal height measuring tape for use in antenatal care in developing countries at village level. Tape measurement of symphysis-fundal height has been weight gain, with mid-upper arm circumference as the suggested as a screening test for the detection of fetal least useful predictor. Lechtig (1988) concluded that: arm growth retardation, macrosomia and multiple pregnancy circumference, independently of gestational age, had the (Westin 1977; Belizan et at. 1978; Quaranta et al. 1981; same sensitivity and specificity as weight gain during Calvert et al. 1982; Cnattingius et al. 1984; Mathai et al. pregnancy for gestational age. He proposed the use of arm 1987; Pearce & Campbell 1987; Neilson et al, 1988). It is and non-invasive circumference with either uterine height or weighi for as a height. In Tanzania, maternal and are procedure. Measures for the prediction of birthweight include symphysis-fundal height, maternal height, premeasured routinely in antenatal services, but not fundal pregnancy weight, weight and mid-upper arm We therefore compared the value of different maternal circumference, but published studies have given conflicting anthropometric indices for predicting birthweight and results. Moller (1988) found the highest sensitivity and specificity for symphysis-fundal height and the o f a p re symphysis fundal height measurement for the prediction of low birthweight and twin pregnancy or macrosomia without calculation of gestational age. This is useful since Correspondence: Dr G. E. L. Walraven, c /o Nijmegen Institute of International Health, Nijmegen University, PO Box 9101, 6500 HB Nijmegen, The Netherlands. gestational age is often not known. Furthermore, if *sty pr , a s al measuring tape for use in 526 G. E. L. WALRAVEN ET AL. antenatal care could be designed, similar to the mid-upper arm circumference tape for children younger than five years of age developed by Shakir (1974). Subjects and methods Statistical methods The %2 test with Yate’s correction was calculated for comparison of proportions. The Statistical Package for the Social Sciences (SPSS PC + ) was used to measure