Original Research Articles A Comparison of Manual and Ultrasound Measurements ofFundal Height David W. Euans, MD; Pamela D. Connor, PhD; Ricardo G. Hahn, MD; Wm. MacMillan Rodney, MD; and Kristopher L. Arheart, EdD flew Orleans, Louisiana, an d Memphis, Tennessee Background. Uterine fundal height has long been used Results. Ultrasound fundal height and manual fundal to assess fetal growth. The common prediction rule height were equivalent (Ps.01, R2 = .92). Regression I states that the distance in centimeters between the pubic analysis in normal weight and obese patients showed I symphysis and the top of the fundus equals the gesta­ that both methods have the same predictive power in tional age in weeks. The correlation holds best between determining fetal age. weeks 18 and 32. This study compares manual and Conclusions. Manual measurements are still a reliable ultrasound measurements of fundal height. and inexpensive means o f evaluating the course o f preg­ Methods. We compared palpated and ultrasound sym- nancy. Ultrasound is needed when there is a size-vs-date physis-fundal height. We then compared both measure­ discrepancy. Ultrasound is also useful for teaching the ments with gestational age as determined by usual ultra­ measurement of fundal height. sound measurements. Patients were statistically separated by weight into two groups: normal and obese. We hypoth­ Key word,s. Anthropometry; ultrasonography, prenatal; esized that manual measurements of fundal height would obesity; symphysis-fundal height measurement. be affected by obesity and race. ( / Fam Pract 1995; 40:233-236) The most accurate clinical predictor o f gestational age and although measuring techniques can be hindered by ex­ date of confinement is a reliably reported date of last trinsic factors, such as maternal habitus3 and bladder full­ menstrual period.1 Research from the past decade sug­ ness.4 Gestational age can be estimated even without a gests that clinical markers, such as last menstrual period reliable date o f last menstrual period, especially when cou­ and symphysis-fundal height, when consistently moni­ pled with other clinical markers.5’6 The accuracy o f man­ tored and applied, can be as accurate as fetal ultrasound in ual fundal height measurements alone, however, has been predicting gestational age.2 Ultrasound, on the other questioned. Issues have included the general imprecision band, is not necessarily a reliable indicator of fetal age of the measurement,7 variability o f measurement tech­ throughout the entire pregnancy. niques,8 problems of predicting intrauterine growth re­ Several timely measurements o f symphysis-fundal tardation,9’10 and differences related to race.11 height can yield an accurate estimated gestational age, The height of the uterine fundus during pregnancy has been used as a predictor of gestational age for many years. In 1906, McDonald12 established the distance from Submitted, revised, December 5, 1994. the symphysis pubis to the umbilicus to be between 12 Smn the Department o f Family Medicine, Louisiana State University, New Orleans and 20 cm. The fundal height between weeks 18 and 32, (O.W.E.); and the Department of Family Medicine (P.D.C., W.M.R.), the UT/ baptist HealtbPlex Family Medicine Residency Program (R .G .H .), UT-AM I/St measured in centimeters from the upper border of the brands Family Practice Residency Program (W .M.R.), an d the Division o f Biosta- pubic symphysis around the curve of the abdomen and units and Epidemiology, Department o f Preventive Medicine, (K.L.A .), University of Tennessee, Memphis. D r Euans was a t UT-AM I/St Francis Family Practice Rcsi- over the top o f the uterine fundus, approximates the ges­ tlncy Program at the time this research was conducted. Requestsfor reprints should be tational age in weeks. Manual symphysis-fundal height oddressed to David W. Euans, MD, Department of Family Medicine, 1542 Tulane bn, New Orleans, LA 70112. measurement is actually used to estimate the size or vol- 0 W95 Appleton & Lange ISSN 0094-3509 The Journal o f Family Practice, Vol. 40, No. 3(Mar), 1995 233 Fundal Height Measurement Euans, Connor, Hahn , etal time o f the uterus, which in turn yields an estimated measurement, observers were not blinded to either man­ gestational age. The measurement becomes less accurate ual or ultrasonographic measurements. We did not test I in the final weeks o f pregnancy (ie, 36 weeks or later) for interrater reliability. because of positional shifts of the fetus.13 Patients were separated by weight into one of two Our research question was whether measuring the groups: normal or obese. Because of the variety of defini symphysis-fundal height manually rather than ultrasono- tions for obesity, two methods were used to identify over graphically would yield different estimates o f gestational weight patients. First, any individual who was 20% over age. We hypothesized that manual measurements of fun­ ideal body weight (IBW) as defined by the Metropolitan dal height would be affected by obesity. Life Scales16 was classified as obese. Measurement ofbodi mass index (BMI) was chosen as the second method of' Methods classification because it has a stronger statistical relation ship with body fat than does relative weight for height.1' Study participants were selected from consecutive obstet­ A BMI value o f greater than 27.3 kg/cm 2 was used as the ric patients presenting to our academic family practice indicator of obesity. BMI was determined by the follow residency center for routine fetal ultrasound between Jan­ ing formula: BMI in kg/cm2=(weight in pounds!, uary and April 1991. One hundred fifty-nine patients (height in inches)2X703.1.18 carrying a single fetus between 11 and 42 weeks’ gesta­ Simple linear regression and analysis of covariance! tional age were selected to participate.14 Measurements were used to analyze the data. Only 12% of the sample for determining fetal age included biparietal diameter, had manual and ultrasound estimates that differed, head circumference, abdominal circumference, and femur by more than 2 weeks. To measure how well manual length. Because the purpose of our study was to compare fundal height measurements approximated ultrasounc ultrasonographic and manual measurements, ultrasound measures, the parameters o f a simple linear regressions was used to locate the true fundal height. This measure­ ultrasound on manual fundal height were tested. If the ment was recorded, along with the fundal height as de­ manual measurements were to perfectly match the ultra termined by traditional palpation. A comparison was sound measurements, the regression equation would be made between palpated and ultrasound symphysis-fundal ultrasound=0.0 + l.OXmanual. Therefore, a test in which height. Patients were statistically subdivided into two the regression equation intercept=0 and the slope=1 de groups: those who were within a normal weight range for termines the accuracy of the approximation. Equation their height and those who were classified as obese. parameters were tested for all data combined and foil Fundal height measurements were obtained for each obese and normal groups as determined by the two critc patient in a uniform manner. Patients were placed on an ria. An analysis of covariance was used to test whether tkj examination table in a supine position with legs fully ex­ regression equations differed between groups according tended and arms to the side. The examining physician to obesity or race. A significant F test o f the interactior palpated the uterine fundus, and the measuring tape was between the covariate (manual fundal height) and group placed face down in an effort to eliminate measurement (obesity or racial) indicates that group affiliation affect bias.15 Measurements were made from symphysis to fun­ the slope; a significant F test o f the group main effec: dus, over the fetal axis, with relaxed abdominal and uter­ indicates that group membership affects the intercept ine musculature. The ultrasound fundal height was deter­ Test results with a significance level o f .05 or less were mined by positioning the transducer at the superior aspect considered statistically significant. of the uterus so that the top o f the fundus was visible on the imaging screen. A finger was placed under the probe Results until its shadow coincided with the uppermost aspect of the uterus. This point was marked on the abdomen. The O f the 159 patients selected for the study, only 119 hit distance from the superior aspect o f the pubic symphysis both manual and ultrasound fundal height measure to this point represented the true fundal height. Fetal age ments. Forty patients did not receive both measurement was then determined by ultrasound using multiple mea­ because o f differences in medical coverage arrangements surements on all subjects. All ultrasound measurements Therefore, all results are based on these 119 patients. were performed by the authors, using an A T L/A D R The regression of ultrasound fundal height on mail 4000 ultrasound scanner with a 3.5-MHz probe. ual fundal height was significant (P <.01) and had anil Fundal height was measured by residents and faculty value o f .92. The results o f the tests for slope=l ait physicians with obstetric care experience ranging from a intercept=0 were not significant (both P=N S), indial few months to more than 15 years. Residents’ measure­ ing that there is no significant difference between man* ments were verified by faculty. When taking a second and ultrasound measurements of fundal height. The rcl 234 The Journal of Family Practice, Vol. 40, No. 3(Mar), 191 Fundal Height Measurement Euans, Connor, Hahn, et al Fundal Height by 2 Measures (cm) Figure 1. Comparison o f manual and ultrasound measurements of fundal height in 119 patients. Figure 2.