Send letter to journal:
Re: Underestimation potentially better than overestimation

Dear Editor,

Luscombe and Owens (1) propose a new formula (weight(kg)=3(age)+7) for accurately estimating weight in the 21st century child. They point out that the classical formula (weight(kg)=2(age+4)) underestimates actual weight in the majority of cases and that this has implications for potentially underdosing of resuscitation drugs and fluids. One of the factors implicated by the authors in this increase in weight is the rising prevalence of childhood obesity.

Changing the prescription of intravenous fluid volumes on the basis of what is essentially an increase in adiposity has wider implications, extending beyond the resuscitation period. Current maintenance intravenous fluid requirements in children are based on the supposition that energy expenditure is linked to body weight, assuming 1ml of water loss is associated with the consumption of 1 kilocalorie (2). In actual fact, resting energy expenditure is more closely related to fat-free mass (3). In addition, it is well recognised that energy expenditure in sick children is significantly lower than in the healthy children studied in the original work on maintenance fluid requirements (4). This leads to the suggestion that current weight-based fluid calculation formulae overestimate maintenance intravenous fluid requirements of children. In addition, blood volume in children is most closely related to lean body mass (5), suggesting that this should be the basis for resuscitation fluid volumes.

A change in weight estimation formula as described might therefore lead to significant, and potentially dangerous, excess fluid being administered to children. It would be interesting to know what proportion of the rise in weight over the last fifty years is related to height increase and therefore might be attributed to an increase in lean body mass. Until there is good evidence of harm, however, the current weight estimation formula should remain in place. It may confer significant benefit, rather than detriment, by virtue of its underestimation.

References:

1. Luscombe M, Owens B. Weight estimation in resuscitation: is the current formula still valid? Arch Dis Child 2007;92:412-5.

2. Holliday MA, Segar WE. The maintenance need for water in parenteral fluid therapy. Pediatrics 1957;19:823-32.

3. Illner K, Brinkmann G, Heller M et al. Metabolically active components of fat free mass and resting energy expenditure in nonobese adults. Am J Physiol Endocrinol Metab 2000;278:E308-15.

4. Hatherill M. Rubbing salt in the wound. Arc Dis Child 2004;89:414-8.

Send letter to journal:
Re: The Formula (Age x 3) + 7, Is A More Accurate Estimate Of Weight In Children.

Dear Editor,

We read with interest the article by Luscombe and Owens (1) who present a potentially more accurate formula (Weight (kg) = (age x 3) + 7) for estimating a child’s weight based on their age derived from the weights of children attending an emergency department at Nottingham. The most familiar method to estimate the weight is based on the formula: Weight = (Age + 4) x 2, as widely taught in the Advanced Paediatric Life- Support (APLS) manual for children aged 1 to 12 years.(2) However this method is known to underestimate the weight, with the difference increasing with age.(3) It is not clear from the literature where this formula originated. The equation crudely describes the relationship between age and the mean weight of children based on the 1990 growth standards.(4) Luscombe and Owens (1) demonstrate that the current formula underestimates the child’s weight by a mean of 18.8% and recommend the implementation of the new formula in children aged 1 – 10 years of age, which underestimates the weight by just 2.48%.

Their conclusions are similar to our findings derived independently in 301 (167 males, 134 females) children, aged 1 – 12 years attending for a general paediatric outpatient appointment at the Sheffield Children’s Hospital. We chose children from an outpatient clinic, as children attending an emergency department are not routinely weighed in their undergarments (children in outpatients are), thus possibly confounding the data.

The mean and median weights for the subgroups were similar (table 1). The traditional formula of Weight = (Age + 4) x 2, underestimates the weight of the children in our sample group at every age, with the percentage difference between actual and estimated weights becoming progressively larger with advancing age. The overall mean difference is 17.3% (median 14.8%) across all the measurements (table 1). The median weights are displayed in Figure 1. The simplest formula which most accurately described the relationship between age and weight (based on the regression equation of the relationship between age and median weight = (3.27 x Age) + 5.26; r2 = 0.96) was similar to Luscombe and Owens(1), Weight = (Age x 3) + 7. This formula tends to overestimate the weight slightly in children aged 3 - 10 years old (2.2% to 8.6%), however the overall mean and median differences between the whole age subgroups was similar at 1.0% and -3.0% respectively (table 1). During puberty the variation in weights increases, probably because of variation in pubertal onset and a reduction in age of pubertal onset over the last decades. Although there is greater variation in weight after puberty, the new formula still gives a better estimate than the current formula.

Paediatric drug and fluid dosing is based on the weight, thus the new formula should optimise therapy in the ill child when a weight is not readily available. While caution is advised in proposing an alternative formula, it may be an appropriate time to consider revising the current formula. Although the authors alluded to the new formula being used for children older than 10 years, they do not present their data. Based on our data we also propose that the new formula be considered for use up to the age of 12 years to ensure consistency with current APLS guidelines. In view of the increase in the population weights and reduction in age of onset of puberty it may be time for considered debate on the adoption of a new formula to estimate a child’s weight.

References:

1. Luscombe M and Owens B. Weight estimation in resuscitation: is the current formula still valid? Arch Dis Child 2007;92:412-415.

2. Advanced Paediatric Life Support. Fourth Edition. 2006; Advanced Life Support Group, BMJ Books, Blackwell publishing.

3. Argall J, Wright N, Macway-Jones K, Jackson R. A comparison of two commonly used methods of weight estimation. Arch Dis Child 2003 88:789- 790.

4. Freeman JV, Cole TJ, Chinn S, Jones PR, White EM, Preece MA. Cross sectional stature and weight reference curves for the UK. 1990. Arch Dis Child 1995;73:17-24.

Table 1 Demographic details, mean and median weights and percent difference between the conventional and new formulas. The actual weight compared to the formula weight (age + 4) x 2; is consistently higher in all age groups. A minus (-) symbol indicates an actual weight that is less than the calculated formula weight.

Figure 1 Median weight for each age group (closed squares) and line of best fit (linear regression plot - broken line) r 2=0.96. (age + 4) x 2 * open squares; (age x 3) + 7 * open triangles.

Send letter to journal:
Re: Estimating children's weights: a developing country perspective

Dear Editor,

We read with interest the study by Luscombe and Owens(1)in which a new age-based formula is proposed for estimating weight in paediatric resuscitation. However, as the authors admit, the study findings are limited to a specific UK population and may not apply elsewhere.

In contrast to the UK, where children's weights are increasing, in Malawi undernutrition is a significant problem with 48% of children under 5 years stunted, 22% underweight for age and 5% wasted (<2SD weight for height/ length)(2). Here and in other developing countries, it seems likely that an age-based formula developed in a well nourished population will over-estimate the weight of many children. As 95% of Malawian children are an appropriate length for their weight, measuring length may offer a better guide to estimating weight.

To test this assumption we reviewed retrospectively data on children aged 1-10 years admitted to the resuscitation area of the Queen Elizabeth Central Hospital Accident and Emergency, Malawi over the last month. We chose these children as they represent those most likely to receive treatment based on an estimated weight. We identified 148 children, of whom 100 had both an age (in months), and weight (measured during admission, and recorded to the nearest 100g). Mean age was 30.3 months (range 1-7 years). Mean weight was 11.1kg (range 6-26kg).

We then used the current formula (2x (age+4)) and the new formula proposed by Luscombe et al (3x age +7) to assess their accuracy against actual weight. The current formula over-estimated weight by a mean of 10.6% (range +28.5 to - 39% where a negative value reflects an over- estimate). The proposed formula over-estimated weight by a mean of 17% (range +21 to - 44%).

Most children requiring resuscitation were under 2 years. In older children (>5 years, n=9), weight was overestimated to a greater extent using the proposed formula, with a mean of 30.9%, compared to an over- estimate of 14% using the current formula. This may reflect the cumulative effect of age on stunting.

The formula recommended by APLS has affected resuscitation training and treatment guidelines worldwide. For example 2x (age +4) has been used to recommend drug doses and fluid volumes for different age groups in the WHO guidelines for the integrated management of childhood illness (IMCI), guidelines used widely in the developing world(3).

It should be remembered that in children with severe malnutrition, a major cause of child mortality worldwide, over-resuscitation is dangerous, potentially precipitating fluid over-load, cardiac failure and death.

We feel it is essential that should a new age-based formula be adopted, especially by such an influential group as APLS, it be made clear that this formula only applies to a UK population and is not appropriate in many other settings.

Where stunting is common, the length of a child may better reflect their weight and we have evaluated a triage length tape previously(4). The length tape has the advantage that doses can be marked on it, reducing calculation error. However for any tool to be useful it needs to be readily available and a simple age-based equation meets this requirement well. WHO may wish to determine an appropriate, simple and validated equation for use in low resource settings.

References:

1.Luscombe M, Owens B. Weight estimation in resuscitation: is the current formula still valid? Arch.Dis.Child. 2007;92;412-415.

2.National Statistics Office. Malawi Demographic and Health Survey 2004, NSO and ORC Macro, Malawi, 2005.

3.Management of the child with a serious infection or severe malnutrition: guidelines for first-referral level in developing countries. WHO Geneva, 2000.