Gestational age assessment
N. Burton Attico
Accurate assessment of gestational age is essential to optimal prenatal and intrapartum care. Documentation of crestational age not only enables the expectant mother, her family and the physician to know when to expect the onset of labor, but also aids in obstetric assessment and management should complications occur.
Gestational age determinations are critical in the assessment and management of patients who present with such conditions as premature labor, intrauterine growth retardation, postdates pregnancy, premature rupture of fetal membranes and isoimmunization. They also play an essential role in management decisions related to pregnancy, such as scheduling of diagnostic amniocentesis or cesarean section, and in the treatment of some maternal diseases, such as diabetes, chronic renal disease and preeclampsia associated with hypertension. The best management of all these conditions is prospective, involving early, accurate determinations and documentation of clinical indicators of gestation (Figure 1).
The “average” pregnancy is conventionally described as lasting 280 days or 40 weeks from the last normal menstrual period (LNMP), with a normal biologic range of 266 to 294 days, or 38 to 42 weeks.’–‘ Because of both the normal biologic range and individual patient variation, only 4 percent of births actually occur on the estimated date of confinement (EDC).
By definition, a birth that occurs prior to 37 completed weeks of gestation is premature (7.5 percent of pregnancies), and a pregnancy that extends beyond 42 completed weeks is postdates (6 to 8 percent of pregnancies). However, the 280-day, or 40-week, gestational period is based on the assumption that ovulation uniformly occurs 14 days after the LNMP, whereas ovulation actually occurs 14 days before the next menses, a time that may not correspond to 14 days after onset of the last menses. Thus, adjustment of dating is needed in patients with menstrual cycles longer or shorter than 28 days. The dating may also need to be adjusted in patients with pregnancy occurring immediately after cessation of oral contraception or in patients with medically induced ovulation.
Making patients and their families aware of the normal variations in pregnancy and the intricacies of pregnancy dating is an important component of prenatal education and can avert misunderstandings surrounding the question “When is the baby due? ” It is helpful to consistently use a more clinical manner of dating (i.e., weeks versus months of pregnancy) when discussing EDC with patients and families.
In recent years, increasing reliance has been placed on ultrasound imaging examinations for gestational dating. An error of many physicians has been to regard these examinations as definitive and to repeatedly alter the EDC to conform with the results. Ultrasonic assessment is not necessarily the best or only way to date a pregnancy but is one of a number of gestational indicators or milestones that can be used together to date a pregnancy objectively. Each indicator has a particular interval or window” during which it can be used effectively and have maximal accuracy in the dating sequence (Table 1). [1-9] No single indicator is reliable in pregnancy dating. As shown in Table 2, many factors may cause inaccuracy. The advantages and disadvantages of the leading indicators are discussed in the following sections.
LAST NORMAL MENSTRUAL PERIOD
The LNMP is the traditional gestational indicator used in pregnancy dating and is the gold standard” against which all other indicators are compared. [1-4,10,11] Ideally, each patient should document the dates of her menses prior to conception. Unfortunately, patients do not always remember the date of the LNMP, or they may present for prenatal care late in the pregnancy, making retrospective dating more difficult.
All women of reproductive age should be encouraged to keep accurate menstrual calendars, and documenting the LNMP should be a routine part of women’s health care at all clinical visits. A detailed menstrual history, with exact inclusive dates and characteristics of the LNMP, should be obtained on the first pregnancy-related visit. Notations should be made of the amount and duration of flow, the use of oral contraceptives before the pregnancy, and any variation in the character of the last menses from the patient’s usual menstrual pattern.
It is especially important to remember that the first ovulation after cessation of oral contraception is frequently delayed from two to six weeks. Another difficulty in determining the LNMP is implantation bleeding, which may be mistaken for menses. 12 Finally, women who ovulate infrequently may have no recollection of a last menses from which to calculate a projected ovulation.
DATE OF CONCEPTION
Conception or ovulation dates are more accurate than the LNMP, with a variation of only one day in determining gestational age. Certain groups of patients may be able to give an exact date of conception.  Infertility patients, especially those with induced ovulation or artificial insemination, may know the exact date of conception, and adding 266 days will give an accurate EDC. Similar situations occur in patients with a history of a single periovulation coitus, in patients who are charting their basal body temperatures, or in those who regularly have mittelschmerz pain associated with ovulation.
POSITIVE PREGNANCY TESTS
A positive pregnancy test can be an important indicator of the EDC if it is performed early in pregnancy. Ninety-five percent of the slide and tube tests (immunoassays) become positive by 42 days from the first day of the LNMP. Monoclonal antibody tests, radioreceptor assay, enzyme immunoassay and radioimmunoassay can give positive results before the anticipated menses would have begun. [5,13,14]
An accurate predictor of the EDC is provided if a negative pregnancy test is followed by a positive pregnancy test within one to two weeks. Generally, a positive pregnancy test is noted on the prenatal record, but this is more useful in estimating gestational age if the record also documents a prior negative test, including the specific test that was performed and the exact date of testing.
Commonly used pregnancy tests are listed in Table3, with their respective sensitivity levels in relation to the time of the missed menstrual period. [5,13,14]
Prior to the 16th week of gestation, uterine size on bimanual pelvic examination is an accurate indicator of gestational age. However, the accuracy of this measurement is a function of the experience and skill of the examiner and depends on the absence of uterine or pelvic pathology. After the first trimester, abdominal displacement of the uterus may cause inaccuracies in this measurement. If uterine size does not correspond with the estimated dates at the time of the initial clinic visit, early sonography is indicated to help determine dates or to better delineate possible complications, such as ectopic pregnancy, molar gestation or pelvic tumors.
From 18 to 35 weeks, the height of the uterus in centimeters closely approximates the gestational age in weeks (McDonald’s rule). For accuracy, this measurement should be taken by tape, not by obstetric calipers. [2,3,10] The measurement should be made from the top of the symphysis pubis over the curve of the uterine corpus to the most cephalic portion of the fundus uteri. This relationship ceases to be linear when the presenting part of the fetus begins to descend into the pelvis (lightening), even though the fetus continues to grow.
Another useful milestone is the date when the uterine fundus reaches the level of the umbilicus, which occurs at approximately 20 weeks of gestation. [3,4,10] Again, the accuracy of this measurement in a normal pregnancy depends on maternal factors (body habitus, posture, obesity, muscle tonus, presence of pelvic masses) as well as the skill and experience of the physician.
The mother’s first perception of fetal movement (quickening) is a traditional subjective indicator of gestational dating. [1,3] In primigravidas, the perception of quickening generally occurs between 18 and 20 weeks of gestation. In subsequent pregnancies, mothers are generally aware of quickening approximately ten to 14 days earlier. Maternal perception of fetal movement is more accurate when it is obtained prospectively rather than retrospectively. It is better to ask patients to anticipate and record fetal movement than to ask them to recall the date of quickening weeks or months after its occurrence.
FETAL HEART TONES
Fetal heart motion can be observed by ultrasound examination as early as the seventh week of gestation. An electronic Doppler stethoscope (Doptone) may detect fetal heart tones (FHT) by eight to 12 weeks. FHT may be heard with the DeLee fetoscope at 18 to 20 weeks. Most obstetricians believe that the first identification of FHT by fetoscope is one of the best retrospective indicators of gestational age late in pregnancy; 20 to 22 weeks should be added to determine the EDC. 
The first detection of FHT by fetoscope may be affected by the skill, the hearing acuity and the degree of persistence of the examiner, the thickness of the maternal abdominal wall and the fetal position. If FHT cannot be heard by fetoscope (DeLee nonelectronic, nonamplified) at 17 to 20 weeks of gestation, the patient should be scheduled for weekly or biweekly visits until fetoscope FHT can be detected.  Many physicians, however, rely on Doppler technology to detect FHT because it is easier than listening with the unaided aural fetoscope.
Ultrasound fetal measurements are an accurate method for predicting gestational age [1,8,11] The appropriate use of ultrasound imaging is to assist with the estimation or confirmation of the EDC, rather than for precise determination of the EDC. Ultrasound is most predictive [+ or -] seven to ten days) when a crown-rump length is obtained between six to 14 weeks of gestation. The biparietal diameter measurement is most predictive ([+ or -] ten to 14 days) when obtained before the 20th week of gestation. Unfortunately, the most common time to obtain this measurement for gestational dating is after 20 to 24 weeks, a time when the margin of error of ultrasonography is beginning to increase.
When obtained early in pregnancy, measurement of femur length by ultrasound can be an accurate predictor of fetal age [+ or -] one week). Late in pregnancy, femur length correlates with total fetal length, and thus is subject to the differences [+ or -] three to four weeks) that occur with genetic variations (large paternal or maternal stature), fetal nutrition (intrauterine growth retardation, macrosomia, dysmaturity) and maternal disease (hypertension, preeclampsia, renal disease, diabetes). 
Once the gestation goes beyond 28 weeks, the reliability of a single ultrasound measurement declines [+ or -]three weeks). Serial ultrasound examinations are more valuable than single examinations.  Linear plotting or graphing to identify a prospective EDC can be done with serial examinations [1,11] (Figure 2). In such plotting, it is important to be aware that dating error is along the horizontal axis (dates) rather than the vertical axis (measurements).
In all normal gestations, the biologic variation is two weeks. Thus, there will be a one- to two-week variation in the EDC, whatever gestational indicators or combination of indicators are employed. Use of an ordinary “pregnancy wheel” at each visit is another method of verifying the age of the gestation. Careful plotting of serial milestones along the wheel provides verification of correct dating and gestational progress.
Management of Conflicting Dates
Questions often arise about management of patients in whom ultrasound, physical examination and menstrual dates conflict. As noted previously, all of the current clinical indicators have variations, and some of the variations can span as much as eight weeks [+ or -] three to four weeks). All of these factors must be considered when evaluating the health of mother and fetus.
Incongruity between size and dates should arouse suspicion of specific clinical problems, depending on the gestational period. Early in pregnancy, multiple gestation, hydatidiform mole, incorrect menstrual history, and genetic or developmental defects should be considered in the differential diagnosis. Late in pregnancy, malposition, intrauterine growth retardation, fetal dysmaturity, genetic or developmental defects, multiple gestation, fetal macrosomia, and either polyhydramnios or oligohydramnios may be the diagnosis. In most cases, early obstetric consultation is strongly advised when a discrepancy is noted between size and dates.
Depending on the course of the pregnancy, a combined or composite predicted EDC should be calculated. [1,4,10] The method that was used should be recorded, and the reliability of this revised assessment against the original EDC based on the LNMP should be determined (Table 1).Judgment must be exercised in “weighting” each of the clinical indicators to arrive at the composite predicted EDC. Generally, the notation of the original EDC should not be changed, obscured or removed, since a worst case scenario” based on the original EDC should be used when deciding whether specific monitoring studies or other obstetric actions (including delivery) are needed to protect the health of the fetus or the mother, or both. In weighting clinical indicators, the physician should give more weight to an indicator obtained during the interval when its predictability was optimal, while also considering what other clinical factors were present at that time.
The traditional obstetric indicators of term gestation, although useful in individual case management, are not as reliable as other pregnancy dating indicators. These indicators, which have wide variation (as much as six weeks), include fetal position and station, lightening, presence or absence of a cervical mucus plug, and cervical effacement, position, firmness and dilation (cervical ripeness). Some of these factors have been combined into the Bishop score, which is widely used as a predictor for labor inductions.  Reasons for Bishop score variations are myriad, and such scores can change overnight, attesting to their inaccuracy in predicting gestational age.
Retrospective assessment of neonatal age, using Dubowitz scoring after birth, is also subject to error [+ or -] two weeks) and/or variations (genetic, neurologic, physical) in fetal development of the individual infant, and is derived from LNMP data. [12,15,16] Other criteria for assessing fetal maturity are listed in Table 4.
Of all the signs listed in Table 4, only amniotic fluid analysis can be used alone in the third trimester without earlier pregnancy observations, monitoring and recording.  However, amniotic fluid analysis, although helpful in predicting the pulmonary maturity of a fetus and its ability to survive without developing hyaline membrane disease, does not establish gestational age.
In considering ultrasonic measurements as a method of determining fetal maturity, combined ratios of different ultrasound values are helpful in verification and formation of a composite score. Ultrasound values to be used include the cephalic index (normal is a biparietal diameter to occipito-frontal diameter ratio of 79 + 4), the femur length to abdominal circumference ratio (normal range: 0.20 to 0. 24), and the head circumference to abdominal circumference ratio (normal range: 1.0 to 1.3). These values vary outside the normal range in cases of diabetes and intrauterine growth retardation.
Early and continuous prenatal care is the key to accurate monitoring and recording of obstetric milestones and gestational dating. Without early, careful establishment of gestational age, serious obstetric problems, such as intrauterine growth retardation, preterm labor and postdatism, cannot be properly evaluated. Early assessments by various techniques usually are more accurate than later determinations. Changing an EDC should be considered a major obstetric decision, although the EDC may be modified during the course of prenatal care to give a combined or composite EDC. Detailed reasons for such alteration should be carefully noted in the medical record. The original menstrual EDC should always be available for clinical consideration in the patient’s subsequent management.
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Indian Health Service, the U S. Public Health Service or the Department of Health and Human Services.
ACKNOWLEDGMENT: The authors appreciate the suggestions and assistance of William Haffner, M. D., Daniel Napoliello, R. N., and Michael Friedman in the preparation of the manuscript and illustrative material.
N. BURTON ATTICO, M.D., M.P.H. is chief of the Maternal Child Health Branch, Phoenix Area Indian Health Service, and assistant chief of the Department of Obstetrics and Gynecology at Phoenix Indian Medical Center. Dr. Attico is a graduate of the Howard University College of Medicine, Washington, D.C., and of the School of Public Health, University of California, Berkeley.
DOROTHY J. MEYER, C. N. M., M. P. H. is a health consultant at the Maternal Child Health Branch, Phoenix Area Indian Health Service. She is a graduate of the College of Nursing, Arizona State University, Tempe, and of the School of Hygiene and Public Health, Johns Hopkins University, Baltimore.
H. JOHN BODIN, M.D. is chief of the Department of Obstetrics and Gynecology at Phoenix Indian Medical Center. Dr. Bodin received his medical degree from Louisiana State University School of Medicine, New Orleans.
DAVID S. DICKMAN, M.D. is deputy chief of the Department of Radiology and director of ultrasonography at Phoenix Indian Medical Center. Dr. Dickman received his medical degree from the Universidad Autonoma de Guadalajara, Mexico.
1. Johnson TRB. Clinical estimation of gestational
age. Contemp Ob/Gyn 1986;28(2):
2. Reed KL. Estimating gestational age. Ariz Perinatal
Soc Newsl 1987;2(3):I-2.
3. Andersen HF, Johnson TR Jr, Barclay ML,
Hora JD Jr. Gestational age assessment. 1. Analysis
of individual clinical observations. Am J
Obstet Gynecol 1981;139:173-7.
4. Haffner WH. Prenatal care. In: Haffner K
Harris, T, eds. Obstetric, neonatal and gynecologic
care. A practical approach for the Indian
Health Service. 5th ed. Rockville, Md.: Ameri
– can College of Obstetricians and Gynecologists
and the Indian Health Service, 1989:BI-3.
5. Derman R, Corson SL, Horwitz CA, et al.
Early diagnosis of pregnancy: a symposium. J
Reprod Med 1981:26(4S):149-78.
6 . Hertz RH, Sokol RJ, Knoke JD, Rosen MG,
Chik L, Hirsch VJ. Clinical estimation of gestational
age: rules for avoiding preterm delivery.
Am J Obstet Gynecol 1978;131:395-402.
7. Rawlings EE, Moore BA. The accuracy of methods
of calculating the expected date of delivery
for use in the diagnosis of postmaturity.
Am J Obstet Gynecol 1970;106:676-9.
8 . Sanders RC. Cephalometry for dating of the
fetus. In: Sanders RC, James AE, eds. The
principles and practice of ultrasonography in
obstetrics and gynecology. 2d ed. New York:
9 . Vorherr H, Messer R. Identification and management
of the post term pregnancy-postmaturity.
In: Zuspan FP, Christian CD, eds.
Reid’s Controversy in obstetrics and gynecology.
3d ed. Philadelphia: Saunders, 1983:
10. Andersen HF, Johnson TR Jr, Flora JD Jr, Barclay
ML. Gestational age assessment. II. Prediction
from combined clinical observations.
Am J Obstet Gynecol 1981;140:770-4.
11. Sabbagha RE, Barton FB, Barton BA. Sonar
biparietal diameter. 1. Analysis of percentile
growth differences in two normal populations
using same methodology. Am J Obstet Gynecol
12. Treloar AE, Behn BG, Cowan DW. Analysis
of gestational interval. Am J Obstet Gynecol
13. Fletcher JL Jr. Update on pregnancy testing.
Prim Care 1986;13:667-77.
14. College of American Pathologists. Interlaboratory
comparison program, diagnostic series,
basic 1987 survey. Skokie, Ill.: College of
American Pathologists, 1988.
15. Sell EJ. Neonatal estimation of gestational age.
Ariz Perinatal Soc Newsl 1987;2(4):I-2.
16. Dubowitz LM, Dubowitz V, Goldberg C. Clinical
assessment of gestational age in the new-born
infant. J Pediatr 1970;77:1-10.
17. AAP Committee on the Fetus and Newborn,
ACOG Committee on Obstetrics: Maternal
and Fetal Medicine. Guidelines for perinatal
care. 2d ed. Elk Grove Village, III.: American
Academy of Pediatrics, 1988:69.
(Tables and other figures omitted)
COPYRIGHT 1990 American Academy of Family Physicians
COPYRIGHT 2004 Gale Group