Advertisement

358: First-trimester ultrasound assessment of gestational age: is earlier better?

      Objective

      Ultrasound (US) dating in the first-trimester has been established to be more accurate than US dating later in pregnancy. However, it is not clear that US dating earlier in first-trimester is more accurate than later in first-trimester. The purpose of this study was to evaluate the accuracy of US assessment of gestational age (GA) in early first-trimester versus late first-trimester in a precisely dated in vitro fertilization (IVF) population.

      Study Design

      All pregnant women with singleton gestations who underwent IVF and presented for an US examination at our medical center between 2005-2008 were evaluated. Patients who underwent US at <11 weeks as well as at 11-14 weeks were included. True GA was determined by the day of ovum retrieval and fertilization minus 14 days. Differences between US estimated GA and true GA were calculated for exams performed at 6-10 weeks gestation and at 11-14 weeks. The accuracy of US estimates at <11 weeks and 11-14 weeks were compared. Paired student's t-test was used in statistical analysis with p<.05 considered statistically significant.

      Results

      272 pregnant women who underwent IVF at our institution had US exams between 11-14 weeks gestation. Of these, 36 women also had exams in our US unit at <11 weeks gestation. US exams performed at < 11 weeks underestimated GA by a mean (± SEM) of 1.17 ± .28 days (p<.001). US exams at 11-14 weeks overestimated GA by a mean (± SEM) +1.59 ± .14 days (p<.001). In 36 patients undergoing measurements at both intervals, the mean (± SEM) absolute value of the discrepancy was 0.31±0.33 days less at 11-14 weeks, though this was not statistically significant (p=.36).

      Conclusion

      US assessment of GA is very accurate throughout the first-trimester. When using first-trimester US to assess the accuracy of last menstrual period (LMP), it is important to consider the timing of the US and the direction of the US dating discrepancy. US estimates at <11 weeks were significantly more likely to underestimate GA and US estimates at 11-14 weeks were systematically more likely to overestimate GA.