Low birthweight in rural Cameroon: an analysis of a cut-off value

The World Health Organisation (WHO) defines low birthweight (LBW) as the weight of a neonate below 2500 g at birth, often corresponding to the 10th percentile for its gestational age [1]. Globally, 20 million neonates are born with LBW, 95% of whom reside in developing countries [1]. Africa, carries the second highest burden of LBW (14.3%) after Asia (18.3%) [1]. LBW babies are about 20-fold more likely to develop complications compared with heavier babies. Also, LBW is a major predictor of mortality among infants in low-income countries, especially during the neonatal period [2]. LBW-associated morbidity is further highlighted as it is a significant determinant of failure to thrive, poor cognitive development and chronic diseases in both childhood and adulthood [3, 4]. Globally, LBW is a good indicator of public health problems like chronic maternal illness, malnutrition, hard-work and poor antenatal care [1]. It has become increasingly evident that the traditional cut-off value of 2500 g advocated by WHO to define LBW may not be generalizable to every settings. For instance, some countries like Sri Lanka with a high incidence of newborns weighing less than 2500 g, do not have an associated high mortality rate [5]. To this effect, WHO has endorsed the definition of a local cut-off value for LBW for every country [1]. However, local data to define LBW in Cameroon, and Africa at large remains meagre. This is worrisome as some babies might not receive appropriate care due to misclassification [6]. Njim and colleagues have conducted the lone study so far, in a sub-urban area, to define a practical cut-off for LBW in Cameroon using the 10th percentile [6]. With the aim of contributing to the available body of evidence on LBW in Cameroon, we conducted this study to: define a clinical cut-off for LBW making comparisons with that provided by Njim et al.; and to identify the predictors and adverse outcomes of LBW babies in a rural setting in Cameroon.

Overall, 1787 records were eligible for this study. Maternal ages ranged from 14 to 49 years, with a mean age of 26 ± 6.5 years. Over three quarters of women in our study were married by the time of delivery. Also, one quarter of the women were delivering for their first time. There was an equal number of male as female neonates (50.6% versus 49.4%; Table 2). The 10th percentile for birthweight was equivalent to 2700 g. There was a moderate level of agreement (kappa = 0.54) with the traditional cut-off value of 2500 g proposed by WHO [1]. However, our cut-off strongly agreed (kappa = 0.997) with the cut-off for LBW reported by Njim et al. in a sub-urban Region of Cameroon [6]. After controlling for maternal HIV status, primiparity, maternal age and gender of the infant (Table 3), preterm delivery was the lone predictor of LBW in multivariate analysis [adjusted odd’s ratio (aOR) = 1.5; 95% CI = 1.1–2.2; p = 0.001; Table 4]. Also, compared with heavier babies, LBW babies were more likely to be stillborn (OR = 9.6; 95% CI = 4.2–21.6; p < 0.001) or suffer from birth asphyxia at the fifth minute of life (OR = 2.0; 95% CI = 1.2–3.3; p = 0.006), Table 5. About 6.1% (163/1787) of babies with birthweights less than 2700 g were found within the grey zone (2500 to < 2700 g). These babies were more likely to be stillborn compared with heavier babies (OR = 6.4; 95% CI = 2.2–18.5; p = 0.006; Table 6).

The locally defined clinical cut-off for LBW in this community was found to be 2700 g, with preterm delivery being the lone predictor of LBW. Also, LBW babies were more likely to be stillborn and asphyxiated at birth. More so, babies within the grey zone were more likely to be stillborn compared with heavier babies.

To the best of our knowledge, only one study in sub-Saharan Africa has defined a local cut-off for LBW [6]. Local cut-off values for LBW have been defined in some developed countries, and ranges from 2750 g in the US in 1992 [8] to 3000 g in Denmark in 2007 [9]. Our cut-off for LBW strongly agreed with the 2600 g previous reported by Njim and collaborators in 2015, in sub-urban Cameroon [6]. These results which are similar suggest that the cut-off for LBW in Cameroon proposed by WHO might not be appropriate for use in clinical care, as vulnerable babies within the grey zone (weights ≥ 2500 g but < 2700 g) are probably not receiving adequate care as they are rather regarded to have a normal birthweight. In this study, 6.1% of the babies were found in this grey area. LBW is a significant predictor for neonatal mortality [10]. Such misclassifications are likely contributors to the overall neonatal mortality. Indeed, babies born with birthweights within this grey zone were about six times more likely to be stillborn compared with heavier babies. This emphasises the need for more studies within the country to harmonise LBW definitions for the institution of appropriate public health policies to address the problem.

Preterm delivery was the lone predictor of LBW in this study. This finding is consistent with other reports from sub-Saharan Africa [6, 11, 12]. It is obvious that preterm infants will have a LBW, as they are not privileged to attain the required term for a normal weight. Preterm deliveries have been associated with poor antenatal care and chronic maternal diseases like HIV [6, 13, 14]. Nevertheless, we found no significant association between being an adolescent, a single mother, a female neonate and LBW. Also, the first pregnancy and delivery, and a positive maternal HIV status did not predict LBW. This finding corroborated with that of Njim et al. [6]. In contrast to observations made by Abubakari et al. [15], we did not find a significant association between primiparity and a female neonate with LBW, probably explained by the fact that the authors did not account for confounders in their study. Finally, LBW was significantly associated with neonatal asphyxia at the 5th minute of life and stillbirth. This finding is consistent with other African series [6, 16]. This reiterates reports by Koum and colleagues which depict LBW as a major contributor to the early neonatal mortality burden in Cameroon [10], and therefore the need to implement evidence-based measures, like defining a local cut-off value for LBW, to provide appropriate care to these infants. Providing care to LBW babies using evidence-based methods might be key in achieving the third goal of sustainable development [17].

Despite its limitations, birthweight remains a relatively reliable parameter to determine the risk of neonatal mortality. This is especially true in rural communities where the use of first trimester ultrasonography for gestational dating is practically non-existent. However, the records of gestational ages (based on the clients’ last menstrual periods) entered into the delivery registers might not be accurate as it is subject to recall bias. The length of the gestation is also crucial in predicting neonatal morbidity and mortality [18], and simply adopting the new cut-off value of 2700 g for the LBW might have limited value in curbing neonatal morbidity and mortality in these rural communities.

We have demonstrated that the clinical cut-off for LBW in this rural community is 2700 g, with preterm delivery being the sole determinant of LBW. LBW babies were more likely to be stillborn or asphyxiated at the fifth minute of life. A significant proportion of these babies may be receiving inadequate neonatal care as the traditional cut-off of 2500 g proposed by the WHO is still used to define LBW in this rural area. However, more data is needed through a prospective study design with larger sample size to define a cut-off value for LBW over the national territory; as this will guide the implementation of effective preventive strategies to curb the morbidity and mortality associated with LBW in Cameroon. While pursuing the validation of a cut-off value for LBW on a larger scale in Cameroon, it is prudent to adopt our current cut-off value for LBW in this rural community.