Original Article
Skin Prick Test Predictive Values for the Outcome of Cashew Challenges in Children

https://doi.org/10.1016/j.jaip.2019.05.057Get rights and content

Background

Cashew is a common cause of tree nut allergy in children. To date there have been few studies of diagnostic tests for cashew allergy, and positive predictive values (PPVs) for cashew as well as other tree nuts are largely extrapolated from studies of peanut allergy. How relevant these cutoffs are for cashew has not been formally explored.

Objective

We aimed to establish skin prick test (SPT) wheal sizes that correlated to 95% PPV for a positive food challenge for cashew.

Methods

We included all cashew oral food challenges (OFCs) conducted as part of the HealthNuts (n = 108; age, 4-6 years) and SchoolNuts (n = 37; age, 10-14 years) studies, both recruited from the community (population cohort). A second cohort of all cashew OFCs conducted at the Royal Children's Hospital (RCH) allergy center (n = 343) (2011-2016) and a private allergy clinic based at RCH (n = 43) was included via electronic medical record review (clinic cohort). The 95% PPV for cashew SPT was calculated for both cohorts.

Results

Among the population cohort (n = 145), 62% of cashew OFCs were positive compared with 20% of the clinic cohort (n = 386). The SPT cutoff for 95% PPV derived from the population cohort was 10 mm (95% confidence interval [CI], 7.5-12.0). For the clinic cohort, the 95% PPV was 14 mm (95% CI, 9.5-unknown). An SPT wheal size of 8 mm had a PPV of 89% (95% CI, 79-95) in the population cohort and 62% (95% CI, 45-78) in the clinic cohort.

Conclusion

A higher SPT wheal size may be more appropriate than the commonly used 8 mm cutoff to guide clinical decisions around when to perform OFC for cashew.

Section snippets

Study populations

The population cohort of this study comprised participants who underwent cashew OFC as part of the HealthNuts and SchoolNuts studies. The HealthNuts and SchoolNuts studies are 2 large population-based allergy studies performed at the Murdoch Children's Research Institute. Both studies recruited participants from the community. Cashew OFCs were performed at 4-year-old and 6-year-old follow-up in the HealthNuts study, whereas the SchoolNuts study assessed students aged 10 to 14 years. The study

Criteria for undergoing cashew oral food challenges

HealthNuts participants with a cashew SPT ≥1 mm or a parent-reported reaction consistent with an IgE-mediated allergy were invited for a clinic appointment with a specialist allergy nurse and OFC. The low SPT cutoff for sensitization was applied in the HealthNuts study to ensure that all potential cases of food allergy were detected. Cashew OFCs were offered at both the 4-year-old and 6-year-old follow-up.

SchoolNuts participants with a parent-reported history of an adverse food reaction

Statistical Methods

Continuously valued variables are summarized using means (and standard deviations) or medians (and range), with frequencies reported as percentages with 95% confidence intervals (CIs) based on the binomial distribution. A 2-sample comparison of the prevalence of positive OFC within the clinic cohorts showed little evidence that these samples were drawn from populations with different prevalence of cashew allergy (P = .24) and similarly for the population-based cohorts (P = .24). There was also

Ethics

Approval to conduct the HealthNuts study was obtained from the Victorian State Government Office for Children (reference no. CDF/07/492), the Victorian State Government Department of Human Services (reference no. 10/07), and the RCH Human Research Ethics Committee (reference no. 27047). For the SchoolNuts study, ethics approval was obtained from the RCH Human Research Ethics Committee (HREC number 31079), the Department of Education, and Early Childhood and the Catholic Education Office. All

Study populations

Demographic and clinical characteristics stratified by the cashew OFC cohort are outlined in Table I. A total of 386 cashew OFCs were performed in a clinical setting (RCH: n = 343, MACCS: n = 43) and 145 cashew OFCs were performed as part of the population-based studies (HealthNuts: n = 108, SchoolNuts: n = 37). Age stratified demographic and clinical characteristics for the clinic cohort are outlined in Table E2 (available in this article's Online Repository at www.jaci-inpractice.org).

Results of cashew OFC

Among

Discussion

To our knowledge, this is the largest series of OFCs for cashew allergy reported and the first study to report 95% PPVs for cashew SPT. We found the SPT threshold for 95% PPV among a clinic cohort of 14 mm and a population cohort of 10 mm with marked differences in cashew allergy prevalence between the clinic and population cohorts.

The strengths of this study are the large number of cashew OFCs and the development of SPT thresholds with 95% PPVs for cashew allergy using data contributed from

Conclusion

Within a population-based cohort we have established a 95% PPV for cashew SPT of 10 mm. It is known that 95% PPVs can vary depending on the population they are generated from and this study has demonstrated considerable variability between clinic- and population-based cohorts for the 95% PPV for cashew SPT. To improve tree nut allergy diagnosis and management, further work is required using data based on OFC-confirmed outcomes with OFC offered to all sensitized individuals irrespective of wheal

References (34)

Cited by (14)

  • Cashew allergy diagnosis: A two-step algorithm leads to fewer oral food challenges

    2022, Journal of Allergy and Clinical Immunology: In Practice
  • Peanut Can Be Used as a Reference Allergen for Hazard Characterization in Food Allergen Risk Management: A Rapid Evidence Assessment and Meta-Analysis

    2022, Journal of Allergy and Clinical Immunology: In Practice
    Citation Excerpt :

    Although patients did not undergo a challenge to peanut, rates of lower respiratory symptoms and/or the need for rescue epinephrine due to reactions across the entire FC dosing range were not greater than those reported in the literature for peanut. Finally, we assessed the rate of anaphylaxis to very low (≤upper 95th CI for the ED05)6 levels of allergen consumption at FC to cashew27,29-36 (Table E1, available in this article’s Online Repository at www.jaci-inpractice.org), hazelnut27-31,37-40 (Table E2, available in this article’s Online Repository at www.jaci-inpractice.org), and walnut29,30,41-43 (Table E3, available in this article’s Online Repository at www.jaci-inpractice.org) reported in the literature, and undertook a meta-analysis (Figures E1-E3, available in this article’s Online Repository at www.jaci-inpractice.org). These data are summarized in Table IV.

  • The Accuracy of Diagnostic Testing in Determining Tree Nut Allergy: A Systematic Review

    2021, Journal of Allergy and Clinical Immunology: In Practice
    Citation Excerpt :

    Study size ranged from 61 to 531. One study included a large population cohort,15 whereas the others were classified as outpatient (n = 3) or retrospective (n = 1). Clinical accuracy of SPTs to determine cashew and pistachio allergy is good, with the AUCs presented ranging between 0.81 and 0.94 for cashew and between 0.85 and 0.89 for pistachio.14,15,20

  • Recent advances in the management of nut allergy

    2021, World Allergy Organization Journal
View all citing articles on Scopus

This study is supported by funding from the National Health and Medical Research Council (NHMRC) of Australia (HealthNuts study ID 1006215 and SchoolNuts study ID 1047396), the Australian Food Allergy Foundation, the Victorian Government’s Operational Infrastructure Support Program, and the NHMRC Centre for Food and Allergy Research (ID 1041420). K. J. Allen, S. C. Dharmage, A.-L. Ponsonby, L. C. Gurrin, J. J. Koplin and R. L. Peters hold National Health and Medical Research Council (NHMRC) awards. V. McWilliam received a PhD scholarship from the NHMRC-funded Centre for Food Allergy Research (CFAR). K. Perrett holds a Melbourne Children's Clinician Scientist Fellowship.

Conflicts of interest: V. McWilliam reports personal fees from speaker honorariums for Aspen Global, Abbott Australasia, and Nestle Health Sciences and personal fees from an Advisory Panel Consultancy for Nestle and Nutricia outside the submitted work. K. J. Allen was previously on the medical advisory board of Before Brands. M. L. K. Tang is on the Nestle Medical Advisory Board Oceania; is a past member of the Danone Nutricia Global Scientific Advisory Board and Medical Advisory Board ANZ; has received lecture fees from Danone Nutricia, Abbott Australasia, and Nestle; is employed by and holds stock options for Prota Therapeutics; had performed consultancy services for Deerfield Consultancy, GLC Consultancy, and Bayer; and has a patent through MCRI. The rest of the authors declare that they have no relevant conflicts of interest.

K. J. Allen is no longer affiliated with her declared institution but the study was conducted while she was employed there.

These authors contributed equally to this work.

The HealthNuts investigators who are not individually named as authors are Terence Dwyer, MBBS, MD, MPH, Adrian Lowe, PhD, Melissa Wake, MBChB, MD, FRACP, FAHMS, and Colin Robertson, MD. The SchoolNuts investigators who are not individually named as authors are Susan Sawyer, MD, George Patton, MD, PhD, Jo Douglass, MD, and Peter Vuillermin, MBBS, FRACP, PhD.

View full text