Fetal macrosomia risk estimation

We disclose that we are the co-inventors of the birth weight prediction algorithms described in the issued US patent entitled ‘Methods, systems, and computer program products for estimating fetal weight at birth and risk of macrosomia’ (US patent number 6695780; issued 24 February 2004) and the Australian patent entitled ‘Estimating fetal birth weight and risk of macrosomia’ (Australia patent number 2003287084; accepted 21 November 2006), as well as similar patents pending in Europe, Canada and Japan, in addition to being shareholders in the US corporation Algorithmic Bioscience Inc. Mazouni et al. have described a nomogram for predicting fetal macrosomia, which is potentially of great utility1. However, the nomogram yields implausible risk estimates. For instance, when two women are of the same weight, the taller woman will have a lower predicted risk of macrosomia. Numerous studies have demonstrated the opposite; after controlling for maternal weight, maternal height is positively correlated with birth weight and macrosomia risk2-4. The nomogram also predicts that Europeans with a body mass index (BMI) at delivery of between 28 and 42 kg/m2 are more likely than not to have macrosomic fetuses. Even though almost half of adult Europeans have BMIs in this range, the predicted rate (> 50%) is nearly an order of magnitude higher than the actual European incidence of macrosomia (< 7%)5, 6. The nomogram yields this and other unrealistic predictions because the authors made errors in both their database selection and analytical methods. They selected their subjects based on suspicion of fetal macrosomia; all had symphyseal–fundal heights > 34 cm. As a result, their sample had an astonishing macrosomia rate of 56%, whereas the populations for whom the nomogram is meant to apply have an average rate of only 8%5, 6. In addition, Mazouni and colleagues used stepwise regression and backward elimination to derive their macrosomia prediction equation. These automated, atheoretical techniques are prone to selecting erroneous prediction models7. Complicating this further is the inclusion in their nomogram of a predictor they found to be non-significant (parity), as well as a physiologically implausible curvilinear relationship, i.e. BMIs of 26 and 46 kg/m2 predict a lower risk of macrosomia than intermediate values. The authors claim that their nomogram yields a positive predictive value of 84%. However, this impressive figure is grossly inflated by the extraordinarily high macrosomia rate in their sample. In general gravid populations—in which the macrosomia rate is typically < 12%—the nomogram would yield a positive predictive value of < 37%. The authors note that their nomogram correctly classified 85% of newborns as either normal weight or macrosomic. They cite our previously published equation that predicts birth weight solely from maternal and pregnancy-specific characteristics, but they fail to note that it classifies fetuses with comparable accuracy (83%)8. More recently, we have developed birth weight prediction models that also incorporate fetal sonographic measurements9, 10. This combined approach is accurate for most ethnicities and can be applied as much as 11 weeks in advance of delivery, whereas Mazouni's nomogram requires that an ultrasound examination be performed within 1 week of delivery. The ability to predict birth weight early in pregnancy enables lifestyle modifications (such as restricted caloric intake and/or increased exercise) that can reduce the likelihood of fetal macrosomia. Our combined approach (available online at http://www.BabyWeightFinder.com) correctly classifies 87% of pregnancies with respect to macrosomia up to 11 weeks before delivery, and it correctly classifies 90% of newborns 3 weeks before delivery—all in normal, low-risk populations with standard rates of macrosomia. Furthermore, our model is appropriate for all patients, regardless of symphyseal–fundal height or any other physical findings. Our approach can be applied even when patient records are incomplete or when some ultrasound measurements are missing (although accuracy is greatest when all data are available). The increased accuracy of our method, combined with its ability to predict term fetal macrosomia early in the third trimester, enables practitioners to confidently adopt a proactive approach towards the management of women carrying macrosomic fetuses. H. Stanislaw*, G. G. Nahum , * Department of Psychology, California State University, Stanislaus Turlock, 801 West Monte Vista Avenue, Turlock, CA 95382, MD, USA, Department of Obstetrics and Gynecology, Uniformed Services University of the Health Sciences, Bethesda, MD, USA

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.