Menu
Menu
Menu
Menu
Background: Accurate gestational age (GA) determination allows correct management of high-risk, complicated or post-date pregnancies and prevention or anticipation of prematurity related complications. Ultrasound measurement in the first trimester is the gold standard for GA determination. In low- and middle-income countries elevated costs, lack of skills and poor maternal access to health service limit the availability of prenatal ultrasonography, making it necessary to use alternative methods. This study compared three methods of GA determination: Last Normal Menstrual Period recall (LNMP), New Ballard Score (NBS) and New Ballard Score corrected for Birth Weight (NBS + BW) with the locally available standard (Ultrasound measurement in the third trimester) in a low-resource setting (Tosamaganga Council Designated Hospital, Iringa, Tanzania). Methods: All data were retrospectively collected from hospital charts. Comparisons were performed using Bland Altman method. Results: The analysis included 70 mother-newborn pairs. Median gestational age was 38 weeks (IQR 37–39) according to US. The mean difference between LNMP vs. US was 2.1 weeks (95% agreement limits − 3.5 to 7.7 weeks); NBS vs. US was 0.2 weeks (95% agreement limits − 3.7 to 4.1 weeks); NBS + BW vs. US was 1.2 weeks (95% agreement limits − 1.8 to 4.2 weeks). Conclusions: In our setting, NBS + BW was the least biased method for GA determination as compared with the locally available standard. However, wide agreement bands suggested low accuracy for all three alternative methods. New evidence in the use of second/third trimester ultrasound suggests concentrating efforts and resources in further validating and implementing the use of late pregnancy biometry for gestational age dating in low and middle-income countries.
This study was carried out at the St. John of the Cross Hospital of Tosamaganga (Iringa, Tanzania), the only Comprehensive Emergency Obstetric and Newborn Care Center in Iringa Rural District. Designated as referral hospital of Iringa Rural District Council, it serves an estimated population of 265 000 inhabitants, handling approximately 2300 deliveries per year. The hospital has a total of 165 beds, 48 of which are in the maternity department, including 12 obstetrics, 18 in vaginal postpartum and 18 in CS postpartum. A labour room, a neonatal resuscitation room and a Neonatal Special Care Unit are also present [10]. All the mother-newborn pairs with complete data on the three different methods of determining GA were included in the study. The agreement in GA estimation between different methods. All data were retrospectively and anonymously collected from hospital charts and did not contain any information that might be used to identify individual patients. Maternal data included: age, weight, BMI, number of pregnancies, mode of delivery, GA by LNMP recall, GA by ultrasound measurement in the third trimester. Neonatal data included: sex, birth weight, APGAR score, GA by NBS and NBS + BW. The GA refers to the duration of time between conception and delivery. The LNMP recall is the difference between the first day of the last menstrual period and the delivery date. A US is defined as of the third trimester when executed at 28 0/7 weeks of gestation and beyond [1]. Late ultrasound GA determination was performed using the INTERGROWTH-21st project estimation method [11]. The NBS consists in a procedure, performed postnatally up to 96 h after birth, that asses physical and neuromuscular maturity of the neonate to determine its gestational age [12]. NBS + BW refers to the NBS adjusted considering birth weight in the score calculation [9]. The US measurement in the third trimester was separately compared with LNMP recall, NBS and NBS + BW. The sample size calculation was based on information from available literature [8]. Assuming a mean difference of 0 weeks with a standard deviation of 3 weeks, a minimum of 64 subjects were required to have an 80% chance of detecting, as significant at the 5% level, an agreement interval of 8 weeks in the Bland-Altman plot. The final sample size was rounded up to 70 subjects (reaching an estimated power of 85%). Sample size calculation was performed using R 4.1 (R Foundation for Statistical Computing, Vienna, Austria) [13]. Categorical variables were summarized as frequency and percentage. Continuous variables were summarized as mean and standard deviation (SD) or median and interquartile range (IQR). The agreement in GA estimation between different methods was assessed using Bland Altman plot (showing mean difference and 95% agreement limits). The correlation between continuous variables was assessed using Pearson correlation coefficient. Inter-rater reliability between the clinicians was evaluated using intra-class correlation coefficient (ICC) in a subsample of 30 newborns with double assessments. All tests were two-sided and a p-value less than 0.05 was considered statistically significant. Statistical analysis was performed using R 4.1 (R Foundation for Statistical Computing, Vienna, Austria) [13].
N/A
