The accurate determination of gestational age (GA) is of great importance in the field of obstetrics, as it can reveal abnormalities in the duration of pregnancy, including premature delivery, as well as pathologies of fetal growth [1] - [5] . The determination of gestational age is typically based on the application of obstetric criteria, which may include the date of the first day of the last menstrual period (LMP), ovulation, early ultrasound dating, measurement of uterine height, and examination of specific amniotic fluid components. However, the use of pediatric clinical scores and newborn electroencephalogram (EEG) may also be employed in this regard [5] - [8] . According to extant research, EEG is regarded as the most reliable method, provided that it is conducted by a duly qualified expert [7] - [10] .

In countries with constrained resources, electroencephalography (EEG) is frequently unavailable. It is common practice among clinicians to estimate gestational age using a combination of methods, including the WILD and/or birth weight, as well as other available obstetric clinical estimates such as uterine height measurement [11] - [14] . The determination of gestational age in these cases remains challenging due to the high prevalence of maternal malnutrition and intrauterine growth retardation [14] - [16] . In the context of adolescent childbearing and the lack of resources in our setting, several challenges are encountered in determining gestational age. These include a lack of awareness, a delayed start to prenatal care, which hinders early ultrasound dating, and a lack of ultrasound services in peripheral health facilities [1] [2] . In such circumstances, the use of morphological maturation scores, such as the Finnstrom and Farr scores, and neuro-morphological scores, including the Dubowitz and Ballard scores, can facilitate the estimation of gestational age [17] - [22] .

Nevertheless, the utilization of neuro-morphological scores appears to be constrained in instances of neurological distress, particularly in extremely premature infants, and is often more challenging to perform. Therefore, in these peripheral structures, the utilization of morphological scores, such as the Finnstrom, to ascertain the term would appear to be a more accessible and straightforward approach for all practitioners, including physicians and delivery room paramedics [15] [22] - [26] .

The Finnström score includes 7 morphological criteria whose rating varies from 1 to 4. It is a valid and relevant method for the assessment of gestational age [23] .

In light of the unavailability of EEG and early ultrasound dating in our context, coupled with the inherent imprecision of the LMP in adolescents, neonatal assessment by the Finnström score may prove invaluable in validating the diagnosis of preterm delivery in adolescent girls. The objective of this study was to assess the accuracy of the Finnström score in newborns of adolescent mothers, where LMP inaccuracy is a particular concern, in order to validate the diagnosis of preterm delivery and to calculate the correlation coefficients between the GA obtained from this score and that of the LMP with the age of the dating ultrasound.

The prevalence of preterm delivery among teenagers was higher according to the Finnström morphological score and early ultrasound dating (32.2% and 37.7%) than that obtained according to the date of the last menstrual period (17.6%), as evidenced in our study. This discrepancy may be attributed to the fact that adolescent females are frequently unable to ascertain the precise date of their last menstrual period due to a lack of education and an absence of motivation to become pregnant. It is important to note that the date of the last menstrual period does not necessarily correspond to the actual date of the menstrual cycle. The date of the last menstrual period is not always a reliable indicator, particularly in the case of teenage mothers.

In reference to the date of the last menstrual period, our findings align with those of Kakudji et al. [29] , who reported a preterm delivery rate of 12.7% in Lubumbashi, Democratic Republic of the Congo (DRC). The physical immaturity of the uterus, which remains hypoplastic, is frequently the underlying cause of preterm delivery in adolescents [30] . Nevertheless, the prevalence of prematurity according to the Finnstrom score is comparable to that of early ultrasound dating (32.2% vs. 37.7%). These findings are consistent with those of Boiro et al. [23] , who reported a prevalence of 21% versus 23.2%. Early ultrasound remains the most reliable examination for determining gestational age (GA). A study conducted in Paris, France, demonstrated that measuring the craniocaudal length during an ultrasound performed between 11 and 14 weeks of gestation allows for an accurate estimation of the date of conception, with a precision of plus or minus five days in 95% of cases [31] .

A study conducted in Cameroon revealed that in 42.7% of cases, the gestational age (GA) calculated based on the date of the last menstrual period (LMP) differed from that estimated using ultrasound dating. The primary factors contributing to this discrepancy in the estimated date of the last menstrual period were forgetfulness, irregularity of the menstrual cycle, and uncertainty expressed by the pregnant woman regarding the date of the last menstrual period provided [32] . In the study by Latis et al. [33] , a comparison of the Dubowitz score with the Finnström score revealed no statistically significant difference between the two clinical scores. In approximately 10% of cases, the Dubowitz score was not utilized due to the critical condition of the neonates. For these reasons, the Finnström score is the most appropriate for routine clinical work.

In the present study, the date of the last menstrual period and the Finnström score were compared with the results of the dating ultrasound. The date of the last menstrual period demonstrated a weak correlation (r = 0.338; p = 0.021) in newborns of adolescent mothers, while the Finnstrom score exhibited a satisfactory correlation (r = 0.828; p < 0.001). The weak correlation observed for the date of the last menstrual period was attributed to the imprecision inherent in this method, particularly in adolescent girls with a low level of education, who constituted the majority of the study population.

Study limitations

One of the primary constraints of our investigation is that the initial ultrasound dating assessment, which is also contingent upon the expertise of the sonographer, was conducted by disparate operators.

It can be concluded that neonatal clinical assessment for the purpose of determining gestational age is of significant importance in the diagnosis of premature delivery in adolescents, who are frequently subject to inaccuracies in the estimation of the date of the last menstrual period. This assessment allows obstetricians to identify instances of term error and refer premature newborns to a neonatology unit for further care. The Finnström morphological score is an accessible, straightforward, and reliable method for determining gestational age. In light of these considerations, we propose the assessment of this score in low-resource settings, where access to EEG and, on occasion, ultrasound is still a challenge.