Arch Gynecol Obstet (2011) 283:999–1004 DOI 10.1007/s00404-010-1510-5 MATERNO-FETAL MEDICINE The significance of fetal sacral length in the ultrasonographic assessment of gestational age Mustafa Ozat • Mine Kanat-Pektas • Tayfun Gungor • Beril Gurlek • Mete Caglar Received: 20 March 2010 / Accepted: 3 May 2010 / Published online: 19 May 2010 Ó Springer-Verlag 2010 Abstract Objective The present study aims to establish a more sensitive nomogram of fetal sacral length measurements, as well as to specify the accuracy of fetal sacral length for the ultrasonographic assessment of gestational age. Methods The present study investigated a total of 2,184 pregnant women who were referred for routine pregnancy follow-up. All of the reviewed women had uncomplicated singleton pregnancies without known structural and chromosomal fetal anomalies. Results A statistically significant linear relationship was established between sacrum length and gestational age [gestational age = 4.49 ? 0.92 9 sacrum length (r = 0.98, R2 = 0.96)]. The rate of increase in sacrum length of fetuses with a gestational age of \28 weeks was formulated as [gestational age = -0.05 ? 1.01 9 sacrum length (r = 0.96, R2 = 0.98)], while the same formula was [gestational age = -0.09 ? 1.32 9 sacrum length (r = 0.94, R2 = 0.96)] for fetuses with a gestational age of C28 weeks. On the other hand, a statistically significant correlation was found to exist between biparietal diameter (r = 0.68, P = 0.001), head circumference (r = 0.590, P = 0.001), femur length (r = 0.719, P = 0.001) and sacrum length (r = 0.696, P = 0.001). However, the same statistically significant correlation exists between abdominal circumference and the other sonographic measurements (r = 0.223, P = 0.375). Conclusions The fetal sacral length appears as an easily acquired and valuable parameter, which directly and M. Ozat
M. Kanat-Pektas (&)
T. Gungor
B. Gurlek
M. Caglar Department of Perinatology, Dr. Zekai Tahir Burak Women Health Research and Education Hospital, Yunus Emre Mah, Dereboyu Sok. No: 71/2 Yenimahalle, 06170 Ankara, Turkey e-mail: minekanat@hotmail.com strongly correlates with gestational age and other biometrical measurements. Therefore, fetal sacral length may be utilized as a complementary tool in both the evaluation of fetal growth and prediction of gestational age. Further research is required to determine the significance of fetal sacral length in prenatal follow-up. Keywords Fetal sacral length
Gestational age
Ultrasonography Introduction Gestational age is the age of a conceptional product, which is often expressed in weeks. Gestational age is calculated by counting from the first day of the last menstrual period or from 14 days before the date of fertilization. The latter method of calculation is based on the assumption that fertilization occurs 14 days after the first day of the last menstrual period, and hence it is also called conceptional age [1]. Gestational age is of high importance in the evaluation of fetal growth and development. However, gestational age calculations according to the last menstrual period can be biased since the average ovulation date varies physiologically. The recalling ability of women may confound the reliability of these calculations as well [2–4]. Although the method of calculating gestational age with the last menstrual period is mostly convenient, other methods have been proposed for more accurate estimation [5]. Obstetric ultrasonography is the most popular of these techniques, because it allows the physicians to examine the fetal body in detail. During the last three decades, ultrasonography of a number of fetal osseous structures have been adopted for the assessment of gestational age and the 123 1000 evaluation of fetal growth and development. The most frequently used morphometric parameters include biparietal diameter, head circumference and femur length [6, 7]. Moreover, nomograms for other osseous structures, such as the mandible, clavicula, scapula, vertebral arch, iliac bone and foot length, have been established [8–10]. Another suggestion is to measure the length of the fetal sacrum, which should be visualized as a part of a routine ultrasonographic evaluation of a fetus. Hence, associated congenital anomalies such as sacral agenesis or sacrococcygeal teratoma can be detected simultaneously [11]. The present study aims to establish a nomogram of fetal sacral length measurements, as well as to specify the accuracy of fetal sacral length for the ultrasonographic assessment of gestational age. Materials and methods The present prospective cross-sectional analysis was approved by the Institutional Review Board and Ethical Committee of Dr. Zekai Tahir Burak Women Health Research and Education Hospital where the study was conducted between January 2009 and January 2010. The present study investigated a total of 2,184 pregnant women who were referred for routine pregnancy follow-up. All of the reviewed women had uncomplicated singleton pregnancies without known structural and chromosomal fetal anomalies. Pregnancies with prenatal risk factors or diseases complicating pregnancy (such as hypertension, diabetes mellitus, polyhydramnios, oligohydramnios and uterine anomalies) were excluded from the study. Also, only women with accurate dates [those with a definite date of the last menstrual period, regular 26- to 30-day cycles, and crown-rump length (CRL) within ±2 SD] were included. The gestational age of the reviewed pregnancies were all determined according to the well-defined date of the last menstrual period. After delivery, each newborn was examined by pediatricians on the first day of life so that any congenital anomaly, developmental problem or growth disturbance could be specified. Thus, any newborn with the aforementioned clinical conditions were excluded. All of the reviewed women underwent a routine ultrasonographic evaluation during which standard measurements of fetal growth parameters including biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC) and femur length (FL) were made. The measurements of BPD and HC were performed on a transverse sagittal plane, which included falx cerebri, cavum septum pellucidum and thalami. Thus, BPD was specified as the distance from the outer edge of the parietal bone to the inner edge of the other. On the other hand, AC 123 Arch Gynecol Obstet (2011) 283:999–1004 was measured at a transverse plane, which included liver and stomach as well as left portal vein within the umbilical region. The FL was recorded when the calipers were placed along the diaphyseal shaft excluding the epiphysis at the longest axis of the image plane. The fetal sacral length was determined according to the method described by Sherer et al. [11]. Therefore, the distance between the distal tip of the spine (namely fifth sacral vertebra) and the anterior superior aspect of the first sacral vertebra was measured by calipers at the sagittal plane. It is well known that coccyx is yet unossified in the third trimester of pregnancy during which this cartilaginous structure is visualized as a thin echogenic plate at the distal edge of the central ossification center of the fifth sacral vertebra. That is why care was maximized to exclude this echogenic plate when fetal sacral length was measured in third trimester pregnancies (Fig. 1). The mentioned technique for the determination of fetal sacral length was adopted for all phases of pregnancy (beginning from the 16th to 40th week). All of the required measurements were obtained by only one certified and experienced sonographer, who was blinded to the gestational age record at the time of sonographic examination. Every parameter was measured three times so that the mean value could be utilized for statistical analyses and intra-observer variance could be calculated. The ultrasonographic equipment used was a Toshiba Power Vision 6000 machine (model: SSA-370A, Toshiba Corporation, Minato, Tokyo, Japan), which had a 3.5 MHz convex probe. Collected data were analyzed by Statistical Package for Social Sciences version 11.5 (SPSS Inc., Chicago, IL, USA). Continuous data were expressed as either mean ± standard deviation, whereas categorical data were reported in numbers or percentages, where appropriate. Shapiro– Wilks test was used to determine the distribution of continuous variables. The associations among the sonographic Fig. 1 The measurement of fetal sacral length (the distance between the arrow tips refers to the length of fetal sacrum) Arch Gynecol Obstet (2011) 283:999–1004 measurements were determined by either Spearman correlation test (if data distribution was normal) or Pearson correlation test (if data distribution was not normal). The effects of gestational age over the biometrical measurements were assessed by linear regression analysis, while Kappa statistical analysis was utilized to specify the intraobserver variations. P \ 0.05 was accepted to be statistically significant. Results Pregnant women who were included in the present study had a mean age of 24.4 ± 4.9 years (range 18–40 years). The average gravidity of the reviewed women were 2.0 ± 1.5 (range 1–5), whereas their average parity was 0.8 ± 1.0 (range 0–4). The gestational age of the investigated pregnancies was found to be 29.5 ± 6.3 weeks on average (range 14–40 weeks). Approximately, 65.4% of the reviewed pregnancies (756 out of 2184) had a gestational age of C28 weeks, while the remaining 34.6% had a gestational age of \28 weeks. The average BPD, HC, AC, FL and sacrum length values were computed to be 73.5 ± 10.4 mm (range 33– 96 cm), 270.3 ± 50.1 mm (range 12–35 cm), 252.6 ± 23.7 mm (range 105–358 mm), 56.5 ± 7.1 mm (range 21–78 mm) and 29.4 ± 6.2 mm (range 16–42 mm), respectively, for all of the pregnancies that were evaluated in the present study. On the other hand, the intraobserver variation was calculated to be 1.1 ± 0.2%, 1.4 ± 0.6%, 1.5 ± 0.9%, 1.2 ± 0.3% and 1.3 ± 0.1% for BPD, HC, AC, FL and sacrum length, respectively. Table 1 shows the increase in fetal sacral length according to advancing gestational age and Table 2 demonstrates how it varies with increasing gestational age. The linear regression equations establish a statistically significant linear relationship between gestational age and the sonographic measurements of BPD, HC, AC, FL and sacrum length (Table 3). Figure 2 refers to the linear regression equation indicating the relationship between gestational age and sacrum length. The rate of increase in sacrum length of fetuses with a gestational age of \28 weeks was formulated as [gestational age = -0.05 ? 1.01 9 sacrum length (r = 0.96, R2 = 0.98)], while the same formula was [gestational age = -0.09 ? 1.32 9 sacrum length (r = 0.94, R2 = 0.96)] for fetuses with a gestational age of C28 weeks. According to Table 4, a statistically significant correlation was found to exist between BPD, HC, FL and sacrum length. However, the same statistically significant correlation is not found between AC and the other sonographic measurements. 1001 Table 1 Gestational age by fetal sacral length Sacral length (cm) Gestational age (weeks) -2 SD Mean ?2 SD 1.4 16.0 16.1 17.1 1.5 16.2 16.5 18.0 1.6 16.3 16.4 19.1 1.7 16.5 17.3 22.0 1.8 16.7 19.2 22.2 1.9 16.9 20.0 23.1 2.0 2.1 17.3 17.6 20.5 21.3 24.2 24.4 2.2 17.8 22.2 25.1 2.3 19.4 23.4 26.2 2.4 20.6 24.6 27.1 2.5 22.7 25.1 30.0 2.6 23.7 26.4 31.0 2.7 24.6 27.0 31.3 2.8 24.8 28.2 32.0 2.9 25.7 29.1 33.1 3.0 26.6 30.5 34.2 3.1 28.7 31.4 35.3 3.2 29.6 32.3 36.2 3.3 31.8 33.1 38.0 3.4 32.6 34.3 39.0 3.5 3.6 33.7 34.6 34.8 36.1 40.0 40.0 3.7 34.8 36.6 40.3 3.8 35.6 38.0 40.4 3.9 36.4 39.3 40.5 4.0 37.5 39.5 40.6 4.1 37.8 39.8 40.7 4.2 39.6 40.0 40.8 Discussion All the osseous components of the spinal column undergo ossification in a much similar manner. That is, one primary ossification center is located at the vertebral body and two others are situated on each side of the posterior neural arch, making three ossification centers for each sacral vertebra [12]. Radiographic studies have shown that the ossification process at the vertebral body is independent of those on the posterior neural arches. Moreover, it has been suggested that there is a significant correlation between crown-rump length and the number of primary ossification centers within the vertebral bones. That is why it has been assumed that the counts of vertebral ossification centers can be utilized to assess fetal age [13]. 123 1002 Arch Gynecol Obstet (2011) 283:999–1004 4,5 Table 2 Fetal sacral length by gestational age Fetal sacral length (cm) 4,0 -2 SD Mean ?2 SD 16 1.2 1.6 1.9 17 1.3 1.7 2.2 18 1.5 1.8 2.2 19 1.6 1.9 2.3 20 1.7 2.0 2.4 21 1.7 2.1 2.4 22 23 1.8 1.9 2.2 2.3 2.5 2.6 24 2.0 2.4 2.8 25 2.2 2.5 2.9 26 2.3 2.6 3.0 27 2.4 2.7 3.0 28 2.5 2.8 3.1 29 2.4 2.8 3.2 30 2.5 2.9 3.2 31 2.6 3.0 3.3 32 2.8 3.1 3.4 33 2.9 3.2 3.5 34 3.0 3.3 3.7 35 3.1 3.4 3.8 36 3.2 3.5 3.9 37 38 3.3 3.3 3.6 3.7 4.1 4.1 39 3.4 3.8 4.2 40 3.5 3.9 4.2 It is well known that the coccyx remains cartilaginous during the third trimester of pregnancy and at birth and that it becomes ossified only after delivery. Hence the appearance of coccyx differs from that of vertebral bones that have already been undergoing ossification. Therefore, the most distal ossification center within the vertebral column is accepted to represent the fifth sacral vertebra. Since the fetal sacrum is an easily identifiable ossifying structure, its length can be measured by means of ultrasonography in the sagittal plane [12, 13]. Besides being used for the estimation of gestational age, the visualization of sacrum also allows the physician to investigate the anatomic integrity of the lower spine and to specify any related congenital Table 3 Regression equations for estimation of gestational age based on ultrasonographic measurements 123 Sacrum length (cm) Gestational age (weeks) 3,5 3,0 2,5 2,0 1,5 1,0 10 20 30 40 50 Gestational age (weeks) Fig. 2 The measurements of fetal sacral length by gestational age abnormalities such as spina bifida, sacral agenesis and sacrococcygeal teratoma [14]. In the present study, the anterior aspect of the ossification center in the vertebral body was selected as the reference point for the measurement of sacrum length. Definitive knowledge of fetal growth and the accurate estimation of gestational age have been shown to be crucial in the management of both normal and high-risk pregnancies. Moreover, the assessment of gestational age is one of the most important aims of ultrasonography in obstetrics. The biometrical measurements of BPD, HC, AC and FL are the most commonly used parameters for the evaluation of fetal growth and gestational age. Especially, the sonographic assessment of BPD and FL has been addressed as the most precise method for the determination of gestational age. However, proper measurement of BPD can be often difficult in case the fetal head is deeply engaged in direct occipitoanterior or occipito-posterior positions, or when the fetus has a different or abnormal cranial shape [15–17]. Similarly, the measurement of FL is a very simple and accurate method, which enables the estimation of fetal gestational age as well as the detection of fetal limb anomalies and growth disturbances. However, it might be difficult to obtain a reliable measurement of FL due to unsatisfactory fetal position or active fetal movement [15–18]. Another issue to be considered is the cumulative influence of factors, which affect the fetal growth so that all standard parameters Regression equation r R2 SE Gestational age = 2.98 ? 0.33 9 biparietal diameter 0.98 0.96 1.56 Gestational age = 4.46 ? 0.52 9 head circumference 0.97 0.95 1.42 Gestational age = 7.82 ? 0.60 9 abdominal circumference 0.94 0.97 1.58 Gestational age = 3.90 ? 0.80 9 femur length 0.96 0.98 1.61 Gestational age = -0.07 ? 1.41 9 sacral length 0.96 0.99 1.48 Arch Gynecol Obstet (2011) 283:999–1004 1003 Table 4 Correlation between ultrasonographic measurements Ultrasonographic measurements Correlation coefficient (r) Statistical significance (P) Biparietal diameter 0.688 0.001* Head circumference 0.590 0.002* Abdominal circumference 0.223 0.375 Femur length 0.719 0.001* Sacral length 0.696 0.001* * P \ 0.05 indicates statistical significance of fetal growth are altered slowly as the pregnancy progresses. Hence, the variability of standard biometrical parameters increases reaching a variation of 7 weeks at term. That is why it becomes less precise to predict gestational age by taking BPD, HC, AC and FL into account during the third trimester [18, 19]. Sherer et al. [11] were the first to define the normal limits of sacral length and demonstrate a high correlation between sacral length, gestational age and other standard measurements of fetal growth. These researchers compared the sacrum length of 506 normal singleton fetuses and 80 singleton fetuses with abnormal growth. Thus, it was shown that the mean sacrum length in millimeters almost equaled the gestational age in weeks. Also, the degree of linear relationship between the gestational age and sacrum length was found to persist during the whole gestational period in both normal fetuses and fetuses with growth disturbance. Later, Pajak et al. [20] investigated a total of 453 pregnant women and documented a strong correlation between the sacrum length and gestational age, as well as between sacrum length and BPD or FL measurements. As a result, fetal sacral length was addressed as an independent predictor of gestational age. Contrarily, Karabulut et al. [21] claimed that sacrum length does not change throughout pregnancy, and the relationship between gestational age and sacrum length is weaker in the third trimester of pregnancy. Their findings were further confirmed by similar data obtained from the anatomical dissection of aborted fetuses. The results of the present study were compatible with those of Sherer et al. and Pajak et al. [11, 21]. Therefore, a consistent linear increase was noted in fetal sacral length as the gestational age increased. Another contradictory finding is the time at which sonographic visualization of sacral ossification centers becomes possible. Sherer et al. [11, 20] demonstrated the presence of sacral ossification centers by at least 16 weeks of gestation, whereas another previously conducted study reported that the sacral ossification centers were present before 15 weeks. On the other hand, Karabulut et al. [21] could observe sacral ossification centers as early as the 14th gestational week. In the present study, sacral ossification centers could be determined by 16 weeks of pregnancy. The aforementioned discrepancies can be attributed to the difference in the resolution power of the ultrasound machine and variability in the personal experience of the sonographers [18]. When compared with the previously published investigations, the present study surveys a relatively larger group of patients and this may provide better results. Nevertheless, the present study lacks the sacral measurements of fetuses with growth disturbance and this leaves no chance to affirm the reliability of fetal sacral length in the prediction of gestational age. The assessment of fetal growth by ultrasonography is usually addressed as the most feasible method for the estimation of gestational age. However, the sonographer may experience problems in acquiring the most appropriate planes for accurate measurements. The fetal sacral length appears as an easily achieved and valuable parameter, which directly and strongly correlates with gestational age and other biometrical measurements. Therefore, fetal sacral length may be utilized as a complementary tool in both the evaluation of fetal growth and prediction of gestational age. Further research is required to determine the significance of fetal sacral length in prenatal follow-up. Conflict of interest statement None. References 1. Engle W (2004) American Academy of Pediatrics Committee on Fetus and Newborn. Age terminology during the perinatal period. Pediatrics 114(5):1362–1364 2. Adegboye AR, Heitmann B (2008) Accuracy and correlates of maternal recall of birth weight and gestational age. BJOG 115(7):886–893 3. Pearl M, Wier ML, Kharazzi M (2007) Assessing the quality of last menstrual period date on California birth records. Pediatr Perinat Epidemiol 21(Suppl 2):50–61 4. Ananth CV (2007) Menstrual versus clinical estimate of gestational age dating in the United States: temporal trends and variability in indices of perinatal outcomes. Pediatr Perinat Epidemiol 21(Suppl 2):22–30 5. Lynch CD, Zhang I (2007) The research implications of the selection of a gestational age estimation method. Pediatr Perinat Epidemiol 21(Suppl 2):86–96 6. Hoffman CS, Messer LG, Mendola P, Savitz DA, Herring AH, Hartmann KE (2008) Comparison of gestational age at birth based on last menstrual period and ultrasound during the first trimester. Pediatr Perinat Epidemiol 22(6):587–596 7. Sherer DM, Sokolovski M, Dalloul M, Khouy-Collado F, Osho JA, Lamarque MD, Abulafia O (2006) Fetal clavicle length throughout gestation: a nomogram. Ultrasound Obstet Gynecol 27(3):306–310 8. Ho TY, Ou SF, Huang SH, Lee CN, Ger LP, Hsieh KS et al (2009) Assessment of growth from foot length in Taiwanese neonates. Pediatr Neonatal 50(6):287–290 123 1004 9. Rahmatullah B, Besar R (2009) Analysis of semi-automated method for femur length measurement from fetal ultrasound. J Med Eng Technol 33(6):417–425 10. Chang CH, Yu CH, Ko HC, Chen CL, Chang FM (2006) Predicting fetal growth restriction by humerus volume: A three-dimensional ultrasound study. Ultrasound Med Biol 32(6):791–795 11. Sherer DM, Abramowicz JS, Plessinger MA, Woods JR (1993) Fetal sacral length in the ultrasonographic assessment of gestational age. Am J Obstet Gynecol 168(2):626–633 12. Wldjaja E, Whitby EH, Paley MN, Griffiths PD (2006) Normal fetal lumbar spine on postmortem imaging. AJNR Am J Neuroradiol 27(3):553–559 13. Gottlieb AG, Galan HL (2008) Nontraditional sonographic pearls in estimating gestational age. Semin Perinatol 32(3):154–160 14. Qi BQ, Beasley SW, Arsic D (2004) Abnormalities of the vertebral column and ribs associated with anorectal malformations. Pediatr Surg Int 20(7):529–533 15. Varol F, Saltik A, Kaplan PB, Kilic T, Yardim T (2001) Evaluation of gestational age based on ultrasound fetal growth measurements. Yonsei Med J 42(3):299–303 123 Arch Gynecol Obstet (2011) 283:999–1004 16. Degani S (2001) Fetal biometry: clinical, pathological, and technical considerations. Obstet Gynecol Surv 56(3):159–167 17. Salomon LJ, Bernard JP, Duyme M, Dorrion A, Ville Y (2003) Revisiting first-trimester fetal biometry. Ultrasound Obstet Gynecol 22(1):63–66 18. Dudley NJ, Chapman E (2002) The importance of quality management in fetal measurement. Ultrasound Obstet Gynecol 19(2):190–196 19. Caughey AB, Nicholson JM, Washington AE (2008) First versus second trimester ultrasound: the effect on pregnancy dating and perinatal outcomes. Am J Obstet Gynecol 198(6):703 20. Pajak J, Heimrath J, Gabrys M, Woyton J (1998) Usefulness of ultrasonographic measurement for fetal sacral length in assessment of gestational age in physiologic pregnancy. Ginekol Pol 69(7):563–569 21. Karabulut KA, Koyluoglu B, Uysal I (2001) Human foetal sacral length measurement for the assessment of foetal growth and development by ultrasonograpy and dissection. Anat Histol Embryol 10:141–146