Oral immunotherapy in food allergy: how difficult to weigh its risks and benefits?

To the Editor: The editorial by Thyagarajan et al1 entitled ‘‘Peanut oral immunotherapy is not ready for clinical use’’ and published in the July 2010 issue of the Journal argued for limiting the clinical practice of oral immunotherapy (OIT) by underlining that several important questions about OITare still unanswered. Quoting the Hippocratic Oath, ‘‘never do harm to anyone,’’ they also underlined that OIT seemed to be generally safe but not without risks. Wasserman et al2 replied to this editorial, stating that OIT for food allergy is safe and effective. Paradoxically, those same authors also reported the death of a child that happened during an OIT formal trial performed in a tertiary pediatric center, adding that it was due to a dosing error. Unfortunately, Wasserman et al did not report any bibliography for that clinical case. They did not give further details about the tertiary center involved or about the trial in which the child took part. Until now, no OIT-related death has been reported in the literature. Thyagarajan et al replied back, apparently denying the occurrence of such a death, and stated that ‘‘it is correct that there have been no fatalities in the studies of OIT up to the present time.’’3 We have recently published a systematic review about OIT,4 and we also have not found any deaths among all the trials published up to now. In fact, OIT in food allergy has been used in Italy for many years,5-7 and several adverse events have been described, 8 but a death caused by OIT has never been signaled. Nevertheless, the letters quoted above put us in doubt: Did the child die, or did he not?We really think it is important to know whether a life-threatening event has happened because of a dosing mistake while performing OIT. Mistakes are common in medical clinical practice, and to knowhow it happened and which type of food was the trigger would be useful. Indeed, if that death happened in a tertiary center, how many errors could happen if OIT comes into routine practice and starts to be performed also in less experienced centers? We agree with Thyagarajan et al1 that OIT can begin to be recommended only after demonstrating that it is ‘‘superior to nonaction (or the current standard of care).’’ For this reason, we believe that it is necessary to know whether any life-threatening event has happened while performing OIT because we are certain that it will help in carefully and safely weighing the risks and benefits of OIT.