1. Antenatal magnetic resonance imaging (MRI) versus ultrasound for predicting neonatal macrosomia: a systematic review and meta-analysis Malin GL (@MalinGemma), Bugg GJ, Takwoingi Y, Thornton J (@jimgthornton), Jones NW

2. #BlueJC is on Twitter and LinkedIn. Join us!Twitter LinkedIn How does #BlueJC work? – Leung E, Tirlapur S, Siassakos D, Khan K. BJOG. 2013 May;120(6):657-60. http://bit.ly/10VaiRZhttp://bit.ly/10VaiRZ For further information: – Follow @BlueJCHost on Twitter@BlueJCHost – Explore our LinkedIn pagehttp://linkd.in/1Cuz8MZhttp://linkd.in/1Cuz8MZ – Explore our blog:http://bluejournalclub.wordpress.com/http://bluejournalclub.wordpress.com/ – See BJOG Journal Club: http://www.bjog.org/http://www.bjog.org/

3. Scenario A midwife referred an African-Caribbean woman at 33 weeks of her first pregnancy because of ‘large for dates’ on abdominal palpation (symphysis fundal height= 37cm). Her oral glucose tolerance test at 28 weeks was normal. She is overweight (body mass index= 27 kg/m 2 ), but has no other risk factors. She has no family history of obstetrics complication. How would you counsel this woman?

4. Background How common is macrosomia in your practice? How do you currently counsel women similar to the one in the scenario?

5. Background Macrosomia is associated with an increased risk of shoulder dystocia and birth trauma, with associated adverse maternal and neonatal outcomes A systematic review published in 2005 found that two- dimensional (2D) ultrasound was an overall poor predictor of fetal macrosomia (Coomarasamy et al, BJOG) Using ultrasound to assess the general antenatal population who are felt to be ‘large for dates’ is not recommended (NICE) Magnetic resonance imaging (MRI) and 3D ultrasound are increasingly used for fetal assessment

6. The Clinical Question What is the accuracy of antenatal 2D, 3D ultrasound and MRI to predict neonatal macrosomia?

7. Structured question (PICOD) ParticipantsWomen with a singleton pregnancy Intervention2D or 3D ultrasound scan or MRI performed in the third trimester to detect fetal macrosomia. Several formulae are used for predicting macrosomia including estimated fetal weight, based on a combination of sonographic fetal measurements, and abdominal circumference alone ComparisonAnother index test (if used) Outcomes (Reference standard) Birthweight >4000 g, >4500 g, >90th or >95th centile Study DesignSystematic review and meta-analysis of diagnostic accuracy studies

8. Figure 1: PRISMA Flow chart Study selection process for systematic review of the diagnostic accuracy of antenatal ultrasound and MRI scan for fetal macrosomia at birth

9. Methods How did the authors assess the quality of individual studies? (also see suggested reading) What were the problems identified by their quality assessment of individual studies? Critically appraise this meta-analysis using the PRISMA checklist. PRISMA checklist

10. Results of meta-analysis for the main tests Comparison of 2D ultrasound EFW (any Hadlock formula >90th centile or >4000 g) and MRI Index test and threshold Reference standard Number of studies* Sensitivity (95% CI) Specificity (95% CI) +ve LR**-ve LR** 2D ultrasound EFW (any Hadlock) >90th centile or >4000 g Birthweight >90th centile or >4000 g 29 (2085/14762) 0.56 (0.49–0.62) 0.92 (0.90–0.94) 7.2 (5.5–9.4) 0.48 (0.42-0.55) 2D ultrasound AC >35 cm Birthweight >90th centile or >4000 g 4 (113/1831) 0.80 (0.69–0.87) 0.86 (0.74–0.93) 5.8 (3.1-10.6) 0.24 (0.16– 0.35) MRI EFW >90th centile or >4000 g Birthweight >90th centile or >4000 g 3 (41/299) 0.93 (0.76–0.98) 0.95 (0.92–0.97) 20.0 (9.6-41.7) 0.07 (0.02– 0.28) Test for difference in sensitivity or specificity of 2D ULTRASOUND EFW, 2D ULTRASOUND AC >35 cm and MRI: P = 0.0002. *(Cases/total number of women); ** +ve LR= Positive Likelihood ratio (95% CI) and –ve LR= Negative Likelihood ratio (95% CI)

11. Figure 2: ROC plot comparing MRI EFW, 2D USS EFW, and 2D USS AC >35cm The solid circles represent the summary sensitivity and specificity for each test The solid line is the 95% confidence interval The dashed lines are the ‘prediction region’- if a new study was done, there is 95% certainty the accuracy estimate (sensitivity and specificity) would fall in this region

12. What does it mean? Take a hypothetical cohort of 1000 pregnant women Prevalence of macrosomia 17% 170 babies will be born with birthweight >90 th centile or >4000 g Of the 170 macrosomic babies: – 2D ultrasound estimated fetal weight (EFW) will miss 75 – 2D ultrasound abdominal circumference (AC) >35 cm will miss 34 – MRI estimated fetal weight (EFW) will miss 12 babies Of the 830 babies without macrosomia: – 2D ultrasound EFW will incorrectly identify 66 as macrosomic – AC >35cm will incorrectly identify 116 as macrosomic – MRI EFW will incorrectly identify 42 babies as macrosomic.

13. Results & Discussion Can you briefly summarise the results of this study as a one-sentence take-home message? Would the results of this study influence your management of the woman in the scenario? How would the results of this study influence your daily practice?

14. Authors’ conclusions MRI volumetry to estimate fetal weight appeared to be much more sensitive than 2D ultrasound EFW for predicting fetal macrosomia. However, these results were based on few studies and small numbers Further research is required to evaluate this technique, including cost-effectiveness, before it can be applied in clinical practice

15. Suggested reading Coomarasamy A, Connock M, Thornton J, Khan KS. Accuracy of ultrasound biometry in the prediction of macrosomia: a systematic quantitative review. BJOG. 2005 Nov;112(11):1461- 6. (Also see DARE summary here)Accuracy of ultrasound biometry in the prediction of macrosomia: a systematic quantitative reviewhere Schünemann HJ, Oxman AD, Brozek J, Glasziou P, Jaeschke R, Vist GE, Williams JW Jr, Kunz R, Craig J, Montori VM, Bossuyt P, Guyatt GH; GRADE Working Group. Grading quality of evidence and strength of recommendations for diagnostic tests and strategies. BMJ. 2008 May 17;336(7653):1106-10. Leeflang MM, Deeks JJ, Takwoingi Y, Macaskill P. Cochrane diagnostic test accuracy reviews. Syst Rev. 2013 Oct 7;2:82.

16. Authors’ Affiliations 1.School of Medicine, the University of Nottingham 2.Department of Obstetrics, Queen’s Medical Centre, Nottingham University Hospitals NHS Trust 3.School of Health and Population Sciences, University of Birmingham, Birmingham, B15 2TT There are no conflicts of interest to declare Corresponding author: Dr Gemma Malin E-mail gemma.malin@nottingham.ac.ukgemma.malin@nottingham.ac.uk Twitter @MalinGemma