Article Text

See original article:

Download PDFPDF

Impact of an early weighing policy on neonatal hypernatraemic dehydration and breast feeding
  1. N P Iyer1,
  2. R Srinivasan1,
  3. K Evans1,
  4. L Ward2,
  5. W-Y Cheung3,
  6. J W A Matthes1
  1. 1
    Department of Child Health, Singleton Hospital, Swansea, South Wales, UK
  2. 2
    Department of Chemical Pathology, Morriston Hospital, Swansea NHS Trust, Swansea, South Wales, UK
  3. 3
    Health Services Research Medical School, University of Swansea, Swansea, South Wales, UK
  1. Dr N P Iyer, Department of Child Health, Singleton Hospital, Swansea SA2 8QA, West Glamorgan, South Wales, UK; iyerprabhu{at}


Aims: To ascertain the effect of a policy of early weighing on the detection and severity of neonatal hypernatraemic dehydration (NHD) and on breastfeeding rates in the short and medium term.

Methods: A policy of weighing infants at 72–96 h was introduced from 1 July 2004. Two time periods were studied: pre-policy and post-policy (18 months each). Babies <28 days of age referred to hospital from the community who, on investigation, had plasma sodium concentrations of >145 mmol/l were identified. Age, plasma sodium concentration, percentage loss of body weight at presentation, breastfeeding rates at discharge and at 8 weeks, and complications due to hypernatraemia or its management were compared between the two groups.

Results: 60 cases of NHD were identified: 23 before and 37 after introduction of the policy. After the policy, there was earlier recognition of NHD (median 3 vs 6 days), lower percentage weight loss (11% vs 15%), smaller increase in sodium (147 vs 150 mmol/l), and higher breastfeeding rate at discharge (73% vs 22%) and 8 weeks (57% vs 22%). All the differences were significant (p<0.01). There was one death in the pre-policy group, and none in the post-policy group.

Conclusions: Weighing babies early coupled with appropriate lactation support resulted in the early recognition of NHD, with less dehydration, less severe hypernatraemia, and higher breastfeeding rates in the short and medium term.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Neonatal hypernatraemic dehydration (NHD) is a potentially serious condition. Cerebral oedema, seizures, thrombosis, gangrene, disseminated intravascular coagulation, intracranial haemorrhage and death have been reported.1 2 The incidence of this condition in Britain is reported to be increasing.3 Inadequate feeding is recognised as the most important cause of NHD. Regular weight monitoring has been used as an objective tool to assess the adequacy of feeding. Weight loss greater than 10% of birth weight or failure to regain birth weight by 10 days are conventional cut-offs for seeking a paediatric opinion. Early weighing and examination by trained healthcare professionals to recognise inadequate feeding are suggested good practice recommendations by the American Academy of Paediatrics and the UNICEF UK Baby Friendly Initiative.4 5 However, evidence is lacking on whether early weighing actually leads to a reduction in the occurrence or severity of NHD. In the UK, there is no uniform policy on the timing of neonatal weight monitoring. Most babies are weighed around the time of blood spot collection for neonatal screening on Guthrie cards (day 7–day 10) or at any time if there are obvious clinical concerns. This practice, we understand, has followed recommendations from Laing and Wong,3 although, from our personal experience, we feel that this may be too late to prevent severe hypernatraemia. Following these concerns and a fatal case of severe hypernatraemia, a policy of weighing babies early (72–96 h after birth) and at 7–10 days (at the time of Guthrie test) was introduced in the Swansea NHS Trust.

The objective of our study was to ascertain the effect of this policy of early weighing on the detection and severity of NHD and on breastfeeding rates in the short and medium term.


Swansea NHS Trust provides maternity care for around 3500 women per year. On 1 July 2004, a comprehensive policy was introduced within the Trust to provide more support to breastfeeding mothers and attempt to reduce the incidence and severity of NHD while improving the breastfeeding rate. The policy included weighing of all babies by a trained midwife at 72–96 h and 7–10 days (at the time of Guthrie test). Breast feeding was observed and infant’s elimination pattern (adequate passage of urine, change of stool from meconium to yellow stools) ascertained to identify mothers who needed extra support with feeding. More support was provided when there was >8% weight loss, and a referral to hospital was made if weight loss exceeded 10% of birth weight. A breastfeeding coordinator was appointed to provide training to midwives and also to provide additional help to mothers, where needed.

We generated a computerised list of subjects using search criteria of <28 days of age and plasma sodium of >145 mmol/l from chemical pathology laboratory archives for 1 January 2003 to 31 December 2005 inclusive. Retrospective case note analysis was performed to obtain clinical information. To ascertain the method of feeding at 6–8 weeks, follow-up health surveillance data were obtained after permission from the child health database held by the community paediatric team. The cohort was divided into two groups: pre-policy and post-policy (18 months each). Outcomes evaluated were age, sodium concentration, and percentage loss of body weight at presentation, breastfeeding rates at discharge and 8 weeks, and any complications due to hypernatraemia or its management. All the weighing measures (Seca weighing scale, model 834), both in hospital and the community, were standardised by the medical physics department of the Swansea NHS Trust according to the manufacturer’s guidelines.

The variables studied were: day of presentation, plasma sodium at presentation, percentage loss of body weight. The data distribution for each variable was studied using the Kolmogorov–Smirnov normality test. The data were found not to be normally distributed (p<0.001 for each parameter). In view of the skewed distribution, the results are presented as median (95% CI). Comparisons of the group medians between the two groups were carried out using the Mann–Whitney U test. We also compared the breastfeeding rates at the time of initial discharge after birth and at 8 weeks between the two groups using the χ2 test. The study was approved by the Clinical Audit Department of Singleton Hospital, Swansea NHS Trust.


Table 1 shows the characteristics of the study groups before and after the introduction of the early weighing policy. The incidence of NHD in the pre- and post-policy groups was 5 and 7.4 per 10 000 live births, respectively. All but one were breast fed at the time of admission. The average maternal age, parity of mother and the mode of delivery, median gestation and median birth weight of the baby, and the date of initial discharge were comparable in the two groups. Most of the mothers in both groups were primiparous. In the pre-policy group, the most common reason for admission was poor feeding. On the other hand, the most common reason for admission in the post-policy group was weight loss in excess of 10%.

Table 1 Characteristics of the patients in the two groups (before and after introduction of the policy of early weighing)

There was no significant difference between the two groups in the initial date of discharge from the hospital after child birth. The median age of presentation after the policy was introduced was 3 days of age, compared with 6 days before the policy (fig 1). Neonates in the post-policy group presented with less severe hypernatraemia and a lower percentage loss of body weight. The proportion of babies presenting with plasma sodium ⩾150 mmol/l was also lower in the post-policy group (18.9% vs 56.5%). The median duration of hospital stay was also significantly lower in the post-policy group. A higher percentage of the babies in the post-policy group was breast fed both at discharge after NHD and at 8 weeks of age. All these results were statistically significant, as shown in tables 2 and 3.

Figure 1 Age at presentation in the two groups (before and after introduction of the policy of early weighing).
Table 2 Differences between neonates with hypernatraemic dehydration before and after the early weighing policy
Table 3 Differences in breastfeeding rates before and after the early weighing policy

None of the children had seizures or any other complications at presentation or during admission in either of the groups. There was one death in the pre-policy group. This child was discharged home on day 3 after birth and readmitted on day 7. At presentation, the child had severe dehydration, plasma sodium of 173 mmol/l, bilateral renal vein thrombosis, and disseminated intravascular coagulation. At the postmortem examination, factor V Leiden heterozygosity was found, which may have contributed to the unfortunate death. There were no deaths or serious morbidity in the post-policy group.


Our study is the first to implement and assess the effect of an early weighing policy on NHD. This resulted in neonates presenting early, with less weight loss, less severe hypernatraemia, and with higher breastfeeding rates at discharge and at 8 weeks of age.

What is already known on this topic

  • Weight monitoring is an objective measure of neonatal hydration and hence successful breast feeding.

  • Monitoring newborn weight does not discourage breast feeding.

What this study adds

  • An early weighing policy coupled with breastfeeding support can reduce the severity of NHD.

  • Breast feeding can be maintained successfully even in mothers whose babies need hospitalisation due to NHD.

Prevention of the growing problem of NHD remains controversial, but weighing newborns has been considered an important tool. It provides objective and clear criteria for referral and identifies mothers who need lactation support. Macdonald et al6 have shown that the pattern of weight loss in breastfed infants is different from that in formula-fed babies. Breastfed babies lost more weight and took longer to regain it than the formula-fed babies. The median time for maximum weight loss was 2.7 days. In many health authorities across the UK, weight is measured around the time of Guthrie card testing (day 7–10). The American Academy of Pediatrics recommends that all breastfeeding newborn infants be seen and weighed by a paediatrician or other knowledgeable healthcare professional at 3–5 days of age.4 Similar recommendations have come from Harding and others from Bristol in the UK.7 A comprehensive policy introduced in the Swansea NHS Trust from July 2004 aimed to detect early lactation failure and to provide more support to lactating mothers. A key element of the policy was to weigh all babies at day 3–4 and subsequently at the time of Guthrie testing. This enabled early detection of NHD in our population. Before the introduction of the policy, we found that babies were being referred for poor feeding on the basis of subjective assessment, and hence were presenting later and with more severe clinical (weight loss) and biochemical (plasma sodium) indices of dehydration.

Clinical detection of dehydration has been shown to be unreliable. In a large study of NHD, Moritz et al8 showed that it is difficult to recognise hypernatraemia in infants on clinical grounds alone. They found that many infants were presenting late with significant hypernatraemia leading to alternative diagnoses and often unnecessary investigations. In their study, although 73% of infants had >10% weight loss, it was not the reason for referral in most cases.

Concerns have been raised that demonstrating weight loss may discourage mothers from breast feeding.9 This potential attrition in breastfeeding rates can be minimised by trained healthcare professionals providing a clear explanation to the mother about the purpose of regular weighing in newborn babies. In a recent population-based study,10 it was shown that monitoring newborn weight as part of a comprehensive policy does not discourage breast feeding. Our study showed that early intervention coupled with more community-based support resulted in more mothers continuing to breast feed. In our experience, NHD undermines the confidence of mothers to breast feed their babies, more so when they are severely dehydrated and requiring prolonged medical intervention. The key lies in the prevention of a severe presentation by early detection and ensuring appropriate professional support so that mothers are not discouraged from breast feeding.

There has been previous research on the utility of early weighing. Macdonald et al6 showed a relation between weight loss and hypernatraemia as a secondary measure, raising the possibility of using early weighing as tool for early detection of NHD. McKie et al10 showed that early weighing leads to higher rates of breast feeding by allowing more targeted support. However, our study is the first to show that early weighing can help to detect NHD early and permit continuation of breast feeding in those who require hospitalisation.

Objective measurements of morbidity and the cost effectiveness were not undertaken in this study. The policy resulted in 1.6 times more hospital admissions of babies with hypernatraemia. In terms of work load to hospital staff, this meant one extra admission because of NHD every 1.3 months. Future larger studies with reproducible results may firmly establish the place of early weighing in newborn health surveillance.



  • Funding: None.

  • Competing interests: None.

Linked Articles

  • Correction
    BMJ Publishing Group Ltd and Royal College of Paediatrics and Child Health