A Comparison of the Accuracy of Various Methods of Postnatal Gestational Age Estimation; Including Ballard Score, Foot Length, Vascularity of the Anterior Lens, Last Menstrual Period and Also a Clinician's Non-Structured Assessment.

Abstract

Gestational age is a strong determinant of neonatal mortality and morbidity. Early obstetric ultrasound is the clinical reference standard, but is not widely available in many developing countries. A prospectively designed diagnostic accuracy study in a tertiary referral hospital in a developing country. Early ultrasound (<20 weeks) was the clinical reference standard. Methods evaluated included anthropometric measurements (including foot length), vascularity of the anterior lens, the New Ballard Score and last menstrual period. Clinicians' non-structured global impression 'End of Bed' Assessment was also evaluated. 106 babies were included in the study. Median age at birth was 34 weeks (interquartile range 29-36). Ballard Score and 'End of Bed' Assessment had a mean bias of -0.14 and 0.06 weeks respectively but wide 95% limits of agreement. The physical component of the Ballard score, the total Ballard score and Foot length's ability to discriminate between term and preterm infants gave an area under the receiver operating characteristics curve of 0.97, 0.96 and 0.95, respectively. Although 'End of Bed' Assessment and Ballard score had small mean biases, the wide confidence intervals render the methods irrelevant in clinical practice. Foot length was particularly poor in Small for Gestational Age infants. None of the methods studied were superior to a non-structured clinician's informal 'End of Bed' Assessment. None of the methods studied met the a priori definition of clinical usefulness. Improving access to early ultrasound remains a priority. Instead of focusing on chronological accuracy, future research should compare the ability of early ultrasound and Ballard score to predict morbidity and mortality. Lay summary. Gestational age describes the time interval between conception and the delivery of the baby. Babies born before 37 weeks of gestation (preterm) or after 42 weeks of gestation (post-dates) have an increased risk of death and specific illnesses. The best way to estimate the gestational age is to perform an ultrasound scan on the mother before 20 weeks. However, this is not widely available in many developing countries. Methods to estimate gestational age after birth include calculating the time from the last period, various measurements of the child (such as weight, foot length or head circumference) physical and neurological markers of maturity and examination of the blood vessels on the lens in the eye. In this study, we assessed how accurate these methods were when compared with the best available method; early ultrasound. We also analyzed the clinicians own personal feeling of what the most likely gestation was, based on an informal 'end of bed' assessment. If a method was to be deemed clinically useful it was agreed that it would have to confidently identify the gestation to within 1 week of the true gestation. None of the methods studied could confidently predict the gestational age of individual babies within 1 week. Ballard scoring and the clinician's informal 'End of Bed' Assessment were the most accurate and also had the smallest inter-operator variability when the results of two separate researchers were compared. Foot length performed particularly badly with babies who were small for their gestational age. None of the methods studied confidently predicted gestational age within a week, so have little use in clinical practice. Access to early ultrasound should be improved. Further research into the relationship between maturity markers such as the Ballard score and the rates of death and specific premature related illnesses is warranted.