Penicillin skin testing in the management of penicillin allergy in an outpatient pediatric population

Abstract

Given the high NPV of the current PST regimen, we do not recommend additional testing with the minor determinant mixture. Despite this high NPV, the utility of PST in the low-risk, low-pretest probability outpatient pediatric population was limited, and select patients may be able to proceed directly to oral challenge.

Frequency of severe reactions following penicillin drug provocation tests: A Bayesian meta‐analysis

“…Included studies were published between 1965 and 2019, and were mostly performed in North America ( n = 46, 41.1%) 21–23,25–27,31,32,34,36–38,40,41,44–46,53,58,69–71,74,75,84,85,88,90,96,99,102,104–107,110–112,114,116,122,123,125–127,131 and Europe ( n = 44, 39.3%) 24,28–30,33,35,39,42,43,47,50,52,55–57,59,61–66,68,73,76–80,82,83,86,91,92,94,97,100,109,118,121,124,128,129,132 . Thirty‐eight studies (34.9%) analyzed exclusively children 23,28,35,36,40,51,56,58–61,63,67,73,74,77,78,82,83,86,87,89,90,95,100,103,106‐110,113,115,118,120,121,130,132 and 28 (25.7%) only analyzed adult patients 21,31,32,37,38,41,49,52,53,62,68,71,76,79,88,94,101,102,104,105,114,117,123–125,127,128,131 . Most studies only included outpatients ( n = 69, 66.3%) …”

“…Most studies only included outpatients ( n = 69, 66.3%) 28–30,32–34,36,39,40,42–45,47–50 ,52–55,57,59–66,68,72 ,73,75,76,78–81,84–86,90–93,95,97–100 ,103,104,107,108,110,111,115,118 ,120–122,124‐126,128–130,132 From the 96 studies reporting information on the timing of the index reaction, 15 studies (15.6%) only evaluated patients reporting immediate allergic reactions, 24,30,33,38–40,42,54,60 ,67,80,91,102,103,130 and 12 studies (10.7%) exclusively evaluated patients reporting non‐immediate index reactions 50,55,56,76,78,79,86,95,98,100,119,121 …”

“…After excluding 2451 records in the screening phase, 351 articles were fully read, of which a total of 112 studies were included in the systematic review 21–50 ,51–100 ,101–132 Of these studies, 108 were included in all general analyses, 21–38 ,40–54 …”

“…Prolonged challenges were performed in 41 (39.4%) of the included studies, 22,26,28,29,31,36–38,41,45,47–50 ,55,56,61,65,68,77,78,80,82,86,89,95,98,100 ,101,108,110,115,118–121,124,128–130,132 while graded challenges were performed in 37 (35.6%) studies, 21,23,24,27,33,35,39,42,51,52,54,57–60,62,63,67,69,72,73,75,76,87,88,90,91,94,97,99,103,106,107,116,122,126,127 and 17 (16.3%) studies opted for single dose DPTs 30,34,40,44,53,70,71,79,84,96,104,105,111,112,117,125,131 . Half of the studies ( n = 59, 52.7%) performed a DPT with the suspected culprit drug 25,33,35,36,39,40,47–51,54–57 …”

Should testing be initiated prior to amoxicillin challenge in children?

“…The pathophysiology of these reactions, when present, is not IgE‐mediated, and hence, skin testing would, by definition, be less accurate in diagnosing them. A retrospective review of 80 paediatric patients evaluated for potential beta‐lactam allergy noted limited utility to skin testing in the evaluation of low‐risk, low–pre‐test probability patients (such as those with delayed exanthems) . A retrospective review of 352 children with a history of a non‐immediate reaction to amoxicillin found the sensitivity and specificity of skin testing in diagnosing these reactions to be 8% and 99.7%, respectively .…”