| Abstract | Cilj istraživanja: ocijeniti ultrazvučno predviđanje novorođenačke mase preinačenom formulom u odnosu na Hadlock 2-formulu, ispitati razliku u odstupanjima procijenjenih masa i mase novorođenčeta te postotak pogrešno predviđenih novorođenačkih masa. Nacrt studije: Istraživanje je provedeno u OB Slavonski Brod (Nova Gradiška) od 1. srpnja 2016. do 31. kolovoza 2020. godine. Istraživanje je prospektivno, kohortno, aplikativno, opservacijsko, kvantitativno te analitičko. Ispitanici i metode: U istraživanju je sudjelovalo 213 punoljetnih trudnica s jednoplodnom trudnoćom koje su bolovale od gestacijskog dijabetesa melitusa i imale tijekom trudnoće porast tjelesne mase veći od preporučenog. Učinjena su ultrazvučna mjerenja biparijetalnog promjera, opsega abdomena i duljine natkoljenične kosti fetusa te procjena fetalne mase Hadlock 2-formulom i preinačenom formulom iz pilot istraživanja unutar tri dana do porođaja (37. i 42. tjedna trudnoće). Uspoređena je preciznost formula. Podaci su analizirani za 133 trudnice, odnosno njihovu djecu. Rezultati: Ukupno je bilo 14,3% trudnica s urednim ITM-om prije trudnoće te 12,0% pretilih trudnica; makrosomne djece bilo je 37,6%. U usporedbi s masom novorođenčeta, razlika je u prosječnom odstupanju u masi procijenjenoj Hadlock 2-formulom (P < 0,001), ali ne u slučaju preinačene formule (P = 0,853). Postojala je razlika u veličini pogreške u procjeni mase između preinačene i Hadlock 2-formule (122,05g vs 191,57g; P < 0,001). Preinačena formula imala je manju srednju vrijednost postotka odstupanja u masi u odnosu na Hadlock 2-formulu (3,11% vs 4,70%; P < 0,001). Učestalost pogreški do 5% odstupanja od novorođenačke mase bila je veća za preinačenu formulu (78,2% vs 60,2%; P = 0,002), a učestalost pogreški od 5% do 10% za Hadlock 2-formulu (33,8% vs 19,5% za preinačenu; P = 0,009). Postotak pogreški ≥10% iznosio je 2,3% za preinačenu i 6% za Hadlock 2-formulu. U slučaju makrosomije, učestalost pogreški do 5% te 5% do 10% iznosila je 80% i 20% za preinačenu formulu te 34,0% i 52,0% za Hadlock-2 formulu (P < 0,001; P = 0,001). Slični rezultati dobiveni su za masu od 3800 grama. Hadlock 2-formulom ispravno je dijagnosticirano 66% makrosomne djece, preinačenom 72%. Specifičnost i preciznost testa bili su neznatno različiti. Pozitivna prediktivna vrijednost preinačene formule iznosila je 81,81%, negativna 97,91%. Zaključak: Preinačena formula pokazala je veću preciznost pri većim masama fetusa. Potreban je veći uzorak u općoj populaciji kako bi se ispitala mogućnost implementacije preinačene formule u svakodnevni rad. |
| Abstract (english) | Objectives: to evaluate the ultrasound prediction of fetal weight by a modified formula in relation to the Hadlock 2 formula. Also, the aim was to examine the difference in weight differences between fetal weights predicted by formulas and newborns weights, and the percentage of incorrectly predicted newborns weights with respect to absolute size error up to 5%, from 5% up to 10% and ≥10%. Study Design: The research was conducted in GH Slavonski Brod (Nova Gradiška) from July 1, 2016 to August 31, 2020. The research is prospective, cohort, applied, observational, quantitative and analytical. Participants and methods: The study involved 213 adult pregnant women with singleton pregnancies who suffered from gestational diabetes mellitus and had weight gain during pregnancy higher than recommended. Ultrasound measurements of biparietal diameter, abdominal circumference, and femur length were performed within three days prior to delivery using Hadlock 2 formula and modified formula from the pilot study. The accuracy of the formulas was compared. All deliveries were between 37 and 42 weeks of gestation. Considering exclusion criteria, the data were analyzed for 133 pregnant women and their children. Results: There were 14.3% of pregnant women with normal BMI before pregnancy, 12.0% of obese pregnant women and 37.6% of macrosomia. Compared with the newborns weights we found a difference in the mean difference of the weights estimated by the Hadlock 2-formula (P < 0.001), but not in the case of the modified formula (P = 0.853). Difference was found in mean values of differences between newborns weights and weights estimated the modified and Hadlock 2 formulas (122.05g vs 191.57g; P < 0.001). Compared to newborns weights, the modified formula had a lower mean value of deviation percentage in weight compared to the Hadlock 2-formula (3.11% vs 4.70%; P < 0.001). The error rate of up to 5% deviation from newborn weight was higher for the modified formula (78.2% vs 60.2%; P = 0.002), and the error rate of 5% to 10% was higher for the Hadlock 2- formula (33, 8% vs 19.5% for modified; P = 0.009). The error rate ≥10% was 2.3% for the modified and 6% for the Hadlock 2 formula. In the case of macrosomia, these error rates of up to 5% and 5% to 10% were 80% and 20% for the modified formula, and 34.0% and 52.0% for the Hadlock 2-formula (P <0.001; P = 0.001 ). Similar results were obtained by analysis with a weight limit of 3800 grams. Hadlock's 2-formula correctly diagnosed 66% of macrosomia, modified formula 72%. The specificity and accuracy of the test were slightly different. The positive predictive value of the modified formula was 81.81%, the negative 97.91%. Conclusion: At higher fetal masses, the modified formula showed greater precision. A larger sample in the general population is needed to examine the possibility of implementing the modified formula into everyday practice. |
