BMJ  2005;331 (2 July), doi:10.1136/bmj.331.7507.0-f

Filler

POEM^*

Anaemia doesn't predict iron deficiency among toddlers

Question Does screening toddlers for anaemia identify those with iron deficiency?

Synopsis An insufficient level of iron (which is used in more than 200 enzymes in the body) is associated with developmental disabilities in young children. Measuring serum haemoglobin as an indicator of anaemia is used to screen for iron deficiency in young children. The author of this cohort study evaluated the correlation between anaemia and iron deficiency by examining the findings of the national health and nutrition survey (NHANES, a stratified population sample performed across the United States), conducted between 1988 and 1994. The survey included 1289 toddlers between the ages of 12 and 35 months, and all of these children underwent complete blood counts, as well as measures of iron stores: ferritin, transferrin saturation, and free erythrocyte protoporphyrin. Iron deficiency, identified in 10.9% of the children studied, was defined as at least two of the iron indices being below normal. Anaemia was defined as a haemoglobin level of less than 11.0 g/dl. There was little relation in this sample between the presence of iron deficiency and anaemia. Children with iron deficiency had an average haemoglobin level of 11.5 g/dl, which, although statistically lower than the average 12.1 g/dl in non-deficient toddlers, was still above the cut-off for anaemia. Only 28% (95% CI 20% to 38%) of toddlers with low haemoglobin actually had iron deficiency. The ability of anaemia to rule out iron deficiency was also low: the sensitivity of the test was only 30% (20% to 40%). In other words, for every 100 toddlers, nine will have anaemia and nine will have iron deficiency, but only three of the children with iron deficiency will be anaemic and only three of the children with anaemia will be iron deficient; not a great overlap. Similar results have been shown in data from New Zealand, Britain, and Europe.

Bottom line These study results present a quandary: We cannot feel assured that a young child doesn't have anaemia if they show a normal haemoglobin level, and we can't be sure that they have anaemia if the haemoglobin level is low. Screening for iron deficiency in toddlers by checking serum haemoglobin misses most children with a deficiency, and most of the children with anaemia do not have an iron deficiency. As the author of this study says, it might make more sense to continue low dose supplementation of iron in all children rather than use a policy of screen and treat.

Level of evidence 1c (see www.infopoems.com/levels.html). All or none randomised controlled trials

---

White KC. Anemia is a poor predictor of iron deficiency among toddlers in the United States: for heme the bell tolls. Pediatrics 2005;115: 315-20[Abstract/Free Full Text].

©infoPOEMs 1992-2003 www.infoPOEMs.com/informationmastery.cfm

^* Patient-Oriented Evidence that Matters. See editorial ( BMJ 2002;325: 983[Free Full Text])

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    StumbleUpon    Technorati    What's this?

Rapid Responses:

Read all Rapid Responses

"Serum haemoglobin" for anaemia screening

Kenneth Campbell
bmj.com, 1 Jul 2005 [Full text]

Aneamia does not predict iron deficiency among toddlers

oscar,m jolobe
bmj.com, 2 Jul 2005 [Full text]

Anaemia predicts a response to iron therapy

Charlotte M Wright
bmj.com, 6 Jul 2005 [Full text]

The response to a therapeutic trial as a diagnostic aid

oscar m jolobe
bmj.com, 11 Jul 2005 [Full text]

Bottom line

John Hain
bmj.com, 27 Jul 2005 [Full text]

Online poll

Find out more>>