Between 1997 and 2010, the
incidence of peanut allergy quadrupled in the US In 2000, the American Academy
of Pediatrics recommended that for children who were inclined toward allergic
diseases, peanuts should be avoided until 3 years of age. Moreover, recommendations
were made to pregnant and lactating mothers to avoid peanut consumption.
Everywhere, from schools to jet aircraft, from restaurants to kiddies’ parties,
the terror of peanuts prevailed. A group of UK allergists noted that among
Jewish children in London, the risk of developing peanut allergy was 10 times
that of Jewish children living in Israel. This was associated with a difference
in the age when peanuts were introduced. Among the London Jewish kids, peanuts
were generally not introduced until the second year of Life. Among Israeli
based Jewish children, it was on average 7 months when peanuts were introduced.
Several small and generally inconclusive studies had suggested that in fact
early introduction of allergens such a cows’ milk or eggs, tended to reduce the
severity of the disease. Thus the UK researchers set out to design a large
randomised trial of the early versus late introduction of peanuts to children.
The trial was called the LEAP trial: (Learning Early About Peanut Allergy[1])
To enter the study, the
infant had to be aged between 4 months and 11 months and to suffer either
severe eczema or severe egg allergy. The children then underwent a skin prick
test for peanut allergy. Of the 834 infants who were screened for the study
(194 did not meet the exact inclusion criteria), 640 underwent randomization.
Of these 542 had shown a negative skin prick test to peanut allergy while 98
did show a positive test. Half of these
groups were then allocated to either complete peanut avoidance or controlled
peanut consumption. Controlled peanut consumption involved a weekly intake of
at least 6 grams of peanut protein from an Israeli peanut snack “Bamba”. The
consumption and avoidance of peanuts would continue for 60 months. The
prevalence of peanut allergy at 60 months was the chosen end point. In
randomised controlled trials, there are two ways to analyse the data. One in
called “intention to treat” and this includes all subjects irrespective of how
they adhered to the protocol. The second is the “per protocol analysis” which
includes only those children that adhered to the study protocol as set out at
the beginning. Taking the former approach first, at 60 months, on average 17%
had peanut allergy and 3% didn’t. When one looked the those who had a positive
skin prick test, a staggering 35% of those who avoided peanuts went on to be
peanut allergic while among those exposed to peanuts, the level was just 11%.
This was thus a 3-fold higher risk. Now taking the per-protocol analysis and
focusing on those who were skin prick positive to peanuts at the outset, 34% of
the avoidance group went on to be peanut allergic (in effect the same as the
intention-to-treat analysis). But among the children who had a positive skin
prick test to peanuts at the outset, and who consumed peanuts, not a single
case of peanut allergy was seen at 60 months.
This trial will have profound
effects on how we treat infant and childhood food allergy. Avoidance is
obviously not the key. Controlled introduction and controlled feeding is the
key but more needs to be done to figure out how this best operates for
different foods and for infants with different allergy experience. However, one
thing is sure. The old default of avoidance is out the window
[1] Du Toit et al (2015). Randomization trial of peanut
consumption in infants at risk for peanut allergy. N Eng J Med 372, 803-813
