Introduction
Birth weight is the single most important factor that determines neonatal outcome and survival [1]. It is a complex character that is determined by genetic, socio-economic and environmental factors [2,3].Gestational age at delivery is the most significant determinant of birth weight [4-6]. Other physiologic factors that can determine birth weight are race, parity, genetics, weight, height, haemoglobin concentration, sex, multiple pregnancies, altitude, nutrition and degree of physical activities [7,8]. Pathological conditions such as antepartum haemorrhage, hypertensive disorders, diabetes mellitus and smoking can also influence birth weight [9,10].
Accurate antenatal fetal weight estimation can be used to predict perinatal outcome [1,2]. It is an important factor to consider when making decisions about time, mode and place of delivery [2,11]. Low birth weight is associated with Intraventricular Haemorrhage, Necrotising Enterocolitis and Respiratory Distress Syndrome [12]. Fetal macrosomia is associated with prolonged labour, injury to the genital tract, vesico-vaginal fistula, increasing rate of cesarean delivery, operative vaginal delivery, uterine rupture and postpartum haemorrhage [13]. Estimated fetal weight is also an important factor in conditions like diabetes mellitus in pregnancy, previous caesarean section, and breech presentation [14-16].
Estimation of fetal weight is done by tactile assessment of fetal size, clinical risk factor assessment methods, use of clinical formulae, use of ultrasound and magnetic resonance imaging [17-19]. All currently available methods for assessing fetal weight have a significant degree of inaccuracy [20]. Some studies have reported that clinical methods of fetal weight estimation are comparable to ultrasound fetal weight estimation, with the advantage of being cheaper and readily available [8,21].
Johnson’s formula is as follows: Fetal Weight in g = (FH (cm) – n) × 155 [67]: Where FH = Fundal Height and n = 12 if the vertex is above the ischial spine or 11 if the vertex is below the ischial spine [22]. If a patient's weight is more than 91kg, 1cm is subtracted from the Fundal Height. Dare’s formula used the product of Symphysis Fundal Height (SFH) and Abdominal Girth (AG) at the level of the umbilicus in cm, and the result was expressed in grams (SFH X AG = fetal weight in grams) [23]. Other clinical formulas for fetal weight estimation are Poulous and Langstadt [24], Johnson and Tosach [25], Dawn formula [26], Ojwang formula [27],Bothner formula [28], and Kongnyuy-Mbu formula [29].
This study aims to determine the mean accuracy of two clinical methods (Dare's and Johnson's) for fetal weight estimation. This is important in low-resource settings where ultrasound scans, constant electricity and manpower may not be readily available. There is limited data on the accuracy of a combined clinical method on the same patient. Dare and Johnson’s method of clinical estimation is simple to teach, reproducible and very objective compared to other clinical methods [30,31].
Methods
The study was a prospective cross-sectional study carried out at the Department of Obstetrics and Gynaecology of the University of Port Harcourt Teaching Hospital in Obio-Akpor local government area, Rivers State, South-South, Nigeria. The hospital serves as a referral centre for neighbouring states in southern Nigeria like Bayelsa, Abia, Imo and Delta states.
The inclusion criteria were women with a singleton pregnancy, cephalic presentation, longitudinal lie, intact membranes, known last menstrual period or Early dating scan on or before 20 weeks’ gestation at term. Term pregnancy was defined as pregnancy of 37 + 0 weeks to 41 + 6 weeks of gestation. Exclusion criteria were polyhydramnios, oligohydramnios, gross congenital anomaly on scan, intrauterine fetal death, malpresentation, uterine fibroids and other pelvic/ intra-abdominal masses, antepartum haemorrhage and those who presented in the advanced first stage of labour. Participants who met the inclusion criteria were randomly selected until the sample size of 144 was achieved. The duration of the study was six months (1st of June 2024 to 30th November 2024).
The interval between the clinical fetal weight estimation and delivery was within 72 hours. If a patient had not delivered within 72 hours of weight estimation, the weight estimation was repeated to obtain a more recent value. The participants emptied their bladder, placed supine, and their SFH and AG were measured to the nearest centimetre using a flexible non-elastic metric tape. Each measurement was taken twice by a different examiner, and the average was taken. Abdominal examination to evaluate the descent of the fetal head into the pelvis was performed immediately. The fetus was considered to be engaged if the fetal head was at two-fifths or one-fifth palpable per abdomen. The estimated fetal weight using Dare’s and Johnson’s formula was calculated in grams using the information obtained. The newborns were weighed within 30 minutes of birth to determine their actual weight using a standard Waymaster analogue scale corrected for zero error.
The analysis of the data was done using the Statistical Product and Services Solutions for Windows® version 21. The Shapiro-Wilk test was conducted on continuous variables to assess data distribution. Since the data met normality assumptions, parametric methods were used. Descriptive statistics were summarised with means and standard deviations. The Pearson Correlation analysis was done to evaluate the impact of age, parity, gestational age, weight, and height on the accuracy of fetal weight estimates using Mean Absolute Percentage Error (MAPE) values. Significance levels were set at a p-value threshold of ≤0.05, with a 95% confidence interval.
Results
Demographic characteristics
Table I provides a comprehensive overview of the key characteristics of the study participants, offering valuable insights into the maternal and gestational attributes of the sample population. The mean maternal age of the participants was 31.59 (±4.8) years. The mean gestational age at the time of assessment was 38.61(±1.2) weeks of gestation.
Fetal Weight Estimates and Actual Birth Weight
Table II provides information about the fetal weight estimates derived from the different clinical formulae and the actual birth weights at delivery. Dare's method of fetal weight estimation had a higher standard deviation from the mean estimated fetal weight.
The findings presented in Table III provide insight into the mean accuracy of the combined clinical methods (Dare's and Johnson's) for fetal weight estimation. The results indicate that these methods consistently overestimate fetal weight, with a mean combined estimate of 4471.65 ± 251.13g compared to the mean actual birth weight of 3612.50 ± 375.07g. This overestimation is further quantified by a Mean Absolute Error (MAE) of 859.15 ± 418.23 g, indicating a substantial average difference between the estimated and actual weights.
Factors associated with the accuracy of fetal weight estimation
The relationship between maternal age and error in the combined estimated fetal weight was not statistically significant. Parity and maternal weight tend to increase the error in the combined estimated fetal weight, while gestational age and maternal height tend to reduce the error in the combined estimated fetal weight. More details are in Table IV.
| Variables | Mean (±SD) |
| Maternal Age (years) | 31.59 (±4.81) |
| Parity | 1.76 (±1.28) |
| Gestational Age (weeks) | 38.61 (±1.19) |
| Maternal Weight (kg) | 86.43 (±14.85) |
| Maternal Height (m) | 1.64 (±0.08) |
| Variables | Mean (±SD) |
| Dare’s Estimate (g) | 4501.07 (±393.97) |
| Johnson's Estimate (g) | 4420.70 (±171.90) |
| Actual Birth Weight (g) | 3612.50 (±375.07) |
| Interval Between Clinical Estimate & Actual Birth Weight (hours) | 8.49 (±7.47) |
| Statistic | Value (± SD) |
| Mean Combined Estimate (g) | 4471.65 (± 251.13) |
| Mean Actual Birth Weight (g) | 3612.50 (± 375.07) |
| Mean Absolute Error (MAE) (g) | 859.15 (± 418.23) |
| Mean Absolute Percentage Error (MAPE) (%) | 25.13 (± 15.18) |
| Proportion of Estimates Within 10% of Actual Weight (10%AW) (%) | 6.94 |
MAE: Combined estimate - Actual birth weight; MAPE: (Error / Actual birth weight) * 100
10%AW=Total occurrences within -10% to +10% range in a list of 144: 10 (10/144) * 100 = 6.94%.
| Pearson’s correlational coefficient [95% CI] | p-value | |
| Maternal age | 0.071 [-0.29-0.75] | 0.398 |
| Parity | 0.367 [0.22-0.41] | <0.001* |
| Gestational age | -0.317 [-0.60-(-2.03)] | <0.001* |
| Maternal weight | 0.387 [0.24-0.55] | <0.001* |
| Maternal height | -0.035 [-37.48-24.61] | 0.682 |
*Statistically Significant (p≤0.05)
Discussion
Findings from this study reveal a significant disparity in accuracy, with the clinical methods consistently overestimating fetal weight. This is evidenced by the MAE of 859.15 grams and the MAPE of 25.13%. This finding aligns with the results of Shittu et al.,[32] 80 who also reported significant overestimation tendencies with clinical methods in South-western Nigeria. Overestimation can lead to unnecessary interventions, such as cesarean sections for suspected macrosomia, adding a potential burden to healthcare systems already constrained by limited resources.
In this study, the combined clinical methods achieved accuracy within 10% of actual birth weight in only 6.94% of cases, which is consistent with findings by Mgbafulu et al.,[29].14 who reported similar figures for the combined clinical methods. These findings highlight potential clinical mismanagement, such as unnecessary cesarean sections, due to overestimation of fetal weight, similarly observed by Yomibo-Sofolahan et al.,[33]. The significant deviation underscores their limited reliability for decision-making in high-stakes scenarios like suspected macrosomia. However, their accessibility and low cost make them valuable in settings where ultrasound technology is unavailable.
This study found that maternal characteristics may influence the accuracy of clinical methods. Maternal weight has a moderate positive correlation (0.387 [0.24-0.55] p<0.001). This indicates that as maternal weight increases, the error in estimation also tends to increase, making the combined clinical methods less accurate. These findings are consistent with those of Yomibo-Sofolahan et al.,[33] who observed that maternal body mass index reduces the accuracy of clinical methods of fetal weight estimation.
Accurate weight estimation is crucial for obstetric decision-making, influencing labor management and delivery outcomes. In resource-limited settings, while clinical methods remain practical due to their accessibility, their tendency to overestimate necessitates cautious application.
Training Programs for Clinical Estimation in Resource-Limited Settings: In settings where ultrasound is unavailable, clinical staff should be trained to recognise the limitations of Dare’s and Johnson’s formulas. Training should focus on using combined clinical methods as preliminary tools, with awareness of their tendency to overestimate, thereby helping to minimise unnecessary interventions based on inaccurate fetal weight estimates.
While this study provides valuable insights into the accuracy of clinical methods for fetal weight estimation, certain limitations must be acknowledged. First, the study was conducted in a single tertiary healthcare facility, which may limit the generalizability of the findings to other settings, particularly rural or resource-constrained environments. Secondly, the sample size, while sufficient for detecting statistical differences between methods, may not fully capture the variability in accuracy across diverse maternal populations or conditions such as multiple gestations and preterm deliveries. Thirdly, the reliance on term pregnancies in low-risk women excludes high-risk groups, which might demonstrate different levels of accuracy for both clinical and ultrasound methods. As such, the findings may not be directly applicable to all obstetric populations.
Conclusion
This study has highlighted the significance of Dare’s formula and Johnson’s formulas for fetal weight estimation. These findings underscore the limitations of clinical methods, which, although accessible and economical, tend to overestimate fetal weight, potentially leading to unwarranted medical interventions, such as cesarean sections for suspected macrosomia. However, in low-resource settings, clinical methods remain practical and essential, despite their tendency to overestimate. Awareness of the limitations associated with clinical methods can help practitioners exercise caution, potentially reducing the risk of unnecessary interventions.
List of abbreviations
AG: Abdominal Girth
FH: Fetal Height
MAE: Mean Absolute Error
MAPE: Mean Absolute Percentage Error
SFH: Symphysis Fundal Height
Declaration
Ethical Clearance
Ethical clearance for this study was obtained from the Ethics Review Committee of the University of Port Harcourt Teaching Hospital on the 27th of July 2023. The protocol number is UPTH/ADM/90/S.11/VOL.XI/1617.
Consent for publication
The participant details are not disclosed, and the ethical approval for publication is covered by the Ethics Review Committee of the University of Port Harcourt Teaching Hospital, with protocol number UPTH/ADM/90/S.11/VOL.XI/1617.
Availability of supporting data
Not applicable
Conflict of interest
The authors have no conflicts of interest to declare.
Funding/ financial support
Research was funded by the authors
Author's contribution
Victor Jinyemiema Abel: Conceptualisation of the study; acquisition of data.
Terhemen Kasso: Conceptualisation and design of the study.
Atochi Prince Woruka: Analysis and interpretation of data; editing of manuscript; and critical review for important intellectual content.
Goody Bassey: Design of the study; critical review for important intellectual content.
Dagogo Abam: Drafting and critical review for important intellectual content.
Acknowledgements
Prof. V.K Oriji and Dr A. Gbenga