Diagnostic accuracy of specific IgE compared to a DBPCFC for detection of cow’s milk allergy

StudyPopulation testedPrevalenceDiagnostic accuracy for cow’s milkComments
Positive test on challengeImmediate reactionDelayed reactionCombined immediate & delayedSensitivity (%)Specificity (%)PPV (%)NPV (%)
Roehr 2001165

EL=DS III
98 children aged 2 months to 11.2 years with atopic eczema (62% mild, 28% moderate and 10% severe). Not stated whether they were suspected of having food allergy.55%
(64% milk, 67% egg, 51% wheat, 16% soya)

(placebo challenge results NR)
49%26%
All atopic eczema
25%
All atopic eczema plus respiratory or gastrointestinal symptoms
84 (any reaction)38 (any reaction)70 (any reaction)59 (any reaction)See atopy patch test

The Pharmacia CAP system was used to measure total and specific IgE levels to cow’s milk, hen’s egg, wheat, and soya (detection limit 35 kU/l). Children were regarded as sensitised if their IgE levels were above the detection limit.

*IgE results for a cut off of 17.5kU/L
85 (immediate reaction)
*22
38 (immediate reaction)
*96
59 (immediate reaction)
*86
71 (immediate reaction)
*54
83 (delayed reaction)
*17
38 (delayed reaction)
*96
48 (delayed reaction)
*75
77 (delayed reaction)
*63
Niggemann 1999166

EL=DS II
107 children aged 5 months to 12 years with persistent moderate-severe atopic eczema, and suspected food-related worsening of eczema or immediate-type clinical reactions by parents and/or referring doctor. ‘usual’ treatments had been used, and ‘no specific diets had been tried in the vast majority of children’ (no further details)51%
(51% milk, 70% egg, 44% wheat, 16% soya)

(placebo challenge results NR)
70%
(64% of milk challenges, 82% of egg, 47% of wheat, 57% of soya)

Of 89% of early reactions manifest as skin reactions, 58% were manifest as eczema, and 23% as combined eczema and urticaria
25%
(28% of milk challenges, 16% of egg, 47% of wheat, 43% of soya)

Of 89% of delayed reactions manifest as skin reactions, 76% were manifest as eczema, and 10% as combined eczema and urticaria
5%
(8% of milk challenges, 2% of egg, 6% of wheat, 0% of soya)

Of 92% of combined reactions manifest as skin reactions, 83% were manifest as eczema, and 17% as combined eczema and urticaria
85386171For at least 5 days before challenge testing children were given a diet of either extensively hydrolysed casein formula (infants and young children) or a few foods diet (older children).

Antihistamines discontinued 72 hours before test. TCS were permitted twice daily (betamethasone 0.01%).
It is not stated whether the eczema was clear/controlled before the test

Food challenges:
Placebo: casein hydrolysate banana flavour solution (128 placebo challenges were undertaken)
Test preparation: 259 food challenges were undertaken, using successive doses of fresh pasteurised cow’s milk containing ultraheated soya milk or fat, raw hen’s egg (white and yolk), and wheat powder. The interval between foods was 30 minutes.
’In general’ two active and one placebo challenge were administered to each child.
Challenges were stopped if clinical symptoms were observed or the maximum dose had been reached.
Children were observed for 48 hours as inpatients. A positive reaction to the food challenge was noted if one of the following occurred: urticaria, angioedema, wheezing, vomiting, diarrhoea, abdominal pain, or exacerbation of eczema. An early reaction was that occurring within 2 hours, and a delayed reaction occurred after 2 hours.

The Pharmacia CAP system was used to measure total and specific IgE levels to cow’s milk, hen’s egg, wheat, and soya (detection limit 35 kU/l). Children were regarded as sensitised if their IgE levels were above the detection limit.

The sequence used to test the foods was determined by the dietician who was not involved in assessing the clinical status of the children during the challenges.

It is assumed that the diagnostic accuracy data refer to any positive test (immediate or delayed response).

The diagnostic accuracy of the history of any food related symptoms was also reported: sensitivity 48% (64% for cow’s milk, 45% egg, 33% wheat, 0 soya) and specificity 72% (100% soy, 74% wheat, 58% milk, 54% egg).
Sampson 1997167

EL=DS III

Related (earlier) publication (including 40 of the children, Sampson 1984168)
196 children and adolescents, aged 0.6–17.9 years with atopic eczema, ‘approximately’ 50% of whom also had asthma and allergic rhinitis. It is not clear whether all were suspected of having food allergy, but some were based on the comments made regarding use of DB and open food challenges.46%
(50% to milk, 73% egg, 49% peanut, 28% soya, 22% wheat, 55% fish)
100%*NRNR1003057100See skin prick test.

The Pharmacia CAP system was used to measure total and specific IgE levels to egg, milk, peanut, wheat, soya, fish. 75% were also tested for inhalant allergens (house dust mite, and cat and dog dander). The detection limit was 35 kU/l; children were regarded as sensitised if their IgE levels were above the detection limit.

The authors also noted that there was no correlation between the level of food allergen-specific IgE and the severity of the allergic reaction.

The authors also investigated the IgE levels that would give 90% and 95% predictive values for each of the six foods tested. [those thresholds giving the most complete results quoted here]
For PPV, the 95% values were:
Egg 6kU/L
Milk 32 kU/L
Peanut 15 kU/L
Fish 20 kU/L (i.e. if a child has a IgE level to fish of 20 or more, they are 95% likely to have a positive reaction on food challenge).
[90% or 95% values not possible for soya or wheat]

For NPV the 90% values were:
Egg 0.6kU/L
Milk 1.0 kU/L
Peanut [not possible]
Fish 5 kU/L (0.9kU/L at 95% value)
Soya 5 kU/L (2 at 95% value)
Wheat 79 kU/L (5 at 95% value)
i.e. if a child has a IgE level to wheat of 79 or less, they are 99% likely not to have a positive reaction on food challenge).
Mehl 2006181

EL=DS III
437 consecutive referrals for suspected food allergy. Children aged 3 months – 14 years (median=13 months). 90% had AE49% (n=341)
(DB and open challenge)
NRNRNR87496279Open challenges were allowed in children <1 year with history of immediate-type reactions. 77% of challenges were double blind.

>1 week elimination diet required before provocation

Eczema was clear before testing.

Hydrocortisone (1%) or betamethasone (0.01%) permitted twice daily. Advice to stop antihistamines 72 hours before provocation

NR=not reported

From: Evidence tables

Cover of Atopic Eczema in Children
Atopic Eczema in Children: Management of Atopic Eczema in Children from Birth up to the Age of 12 Years.
NICE Clinical Guidelines, No. 57.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2007 Dec.
Copyright © 2007, National Collaborating Centre for Women’s and Children’s Health.

No part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK [www.cla.co.uk]. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.