Developing a quality assurance program for transvaginal cervical length measurement at 18-21 weeks' gestation

Background Preterm birth is the leading cause of death in children under the age of five years. Transvaginal cervical length (TVCL) assessment can be used to predict preterm delivery risk at the mid-trimester scan. To optimise the screening tool, developing and maintaining quality standards is important. Aims To develop an Australian reference range for TVCL at 18.0-21.0 weeks' gestation, quality standards for measurement and audit mechanisms for ultrasound operators. Materials and Methods A retrospective audit was performed of consecutive patients scanned at 18.0-21.0 weeks' gestation. Each TVCL measurement ultrasound image was reviewed, and exclusions were made based on a defined set of quality criteria. Fractional polynomial Bayesian methodology was used to establish a reference range. Central tendency, dispersion plots and cumulative sum charts for operators in the original reference range cohort were created. These plots were then applied to a second validation cohort of operators to establish the efficacy of this quality assurance audit tool. Results Median TVCL from 1031 participants was 36.0 mm (interquartile range 32.7-40.0 mm), which was independent of gestational age. The quality audit tool was applied to 15 operators from the reference cohort with a mean cervix length multiple of the median of 1.01 and a mean SD log10 cervix length multiple of the median of 0.06. Of the 22 operators in the validation cohort, 20 (90.9%) demonstrated ideal or acceptable central tendency results, and 19 (86.4%) remained in the appropriate cumulative sum zone. Conclusion An Australian cervix length measurement reference range at 18.0-21.0 weeks' gestation has been developed along with a validated quality assurance audit tool for ultrasound operators.