Kiwi Allergy in Singaporean Children

Since the first report of kiwi allergy in 1981, [1] the kiwi fruit (Actinidia deliciosa) has been increasingly recognised as a common food allergen [2]. Kiwi allergy has been reported in Western and Middle Eastern countries (e.g., Mexico, Europe, Turkey, Israel and Australia) whereas in Asia, the only study from Taiwan reports kiwi as the second most common fruit allergy after mango [2, 3]. There is limited paediatric data regarding kiwi allergy [4]. In Singapore, kiwi is the most common cause of fruit allergy and fruit-induced anaphylaxis [5]. Hence, we aim to describe the clinical features of kiwi allergy in Singaporean children. Using a seven-year retrospective chart review (2017–2024), children aged 0 to 18 years with prick-to-prick test (PPT) performed to kiwi fruit were identified from our departmental electronic database at KK Women's and Children's Hospital. PPTs to raw (green and/or yellow) kiwi fruit were performed. A positive PPT was taken as a mean wheal size of ≥ 3 mm compared with negative control. Diagnosis of kiwi allergy was made based on a convincing history of an immediate allergic reaction and a positive PPT to raw kiwi fruit. The study with waiver of informed consent was approved by SingHealth Centralised Institutional Review Board (reference number 2022/2309). A total of 94 patients were identified. Positive PPT was recorded in 56 (59.6%) patients, of which 51 (54.3%) were diagnosed with IgE-mediated kiwi allergy and 5 (5.3%) were considered sensitised with no known kiwi exposure. Negative PPT was recorded in 38 (40.4%) patients, of which 11 (11.7%) reported clinical tolerance (kiwi tolerant), 5 (5.3%) with no prior exposure to kiwi, and 23 (23.4%) had a history of suspected reaction, whom were either recommended reintroduction or continued avoidance. The eventual outcomes of these 23 patients were not captured because of loss to follow up. Of the 51 kiwi-allergic patients, 22 (43.1%) had a positive PPT to green kiwi, 10 (19.6%) to yellow, 5 (9.8%) to unspecified kiwi and 14 (27.5%) to both green and yellow. The median PPT size was 8.0 mm (interquartile range 5.0–14.0 mm) and 9.0 mm (IQR 5.4–10.6 mm) for green and yellow kiwi respectively. The clinical features of the study population are summarised in Table 1. Cutaneous symptoms were observed in the majority (82.4%) while isolated oro-pharyngeal symptoms were only noted in 11.8%. Anaphylaxis was reported in 4 (7.8%) cases. Age of first visit, in years Median (min–max) 4.1 (0.8–17.4) 1.8 (0.6–9.5) Age of first reaction, in years Median (min–max) 2.3 (0.5–15) Cutaneous Isolated cutaneous involvement 42 (82.4) 27 (51.9) Oro-pharyngeal Isolated oro-pharyngeal 11 (21.6) 6 (11.8) Gastrointestinal Isolated gastrointestinal 12 (23.5) 1 (2.0) Respiratory Isolated respiratory involvement 4 (7.8) — Kiwi allergy has been most commonly described in association with Pollen Food Allergy Syndrome (PFAS) because of its cross-reactivity with birch pollen (Bet v 1), in Latex-Fruit Syndrome (LFS) owing to its cross-reactivity with natural rubber latex allergen (Hev b 6) as well as Lipid Transfer Protein (LTP) syndrome with LTP being pan-allergens present in various foods and plants [2]. As such, the severity of kiwi-allergic reactions ranges from mild oral allergy syndrome in PFAS, to life-threatening anaphylaxis commonly seen in LTP syndrome. Systemic reactions ranging from mild (urticaria) to severe (anaphylaxis) can also occur in kiwi allergy, in the absence of cross-sensitisation. Act d 1 represents the major allergen and is frequently found in mono-sensitised patients who are not allergic to pollen or other fruits (‘mono-allergy’) [4]. In the EuroPrevall study, one of the largest studies to date, among the 311 patients, 63%–72% had allergies to grass/ tree pollen, 2%–11% latex allergies and 5%–18% were mono-allergic to kiwi [3]. In another large study performed in the UK, 29% of patients had allergies to grass pollen, 23% tree pollen, 9% latex, 5% avocado, 6% banana, 6% apple while half were mono-allergic [6]. Our study is the largest report of kiwi-allergic children in Asia, with the majority of children (82.4%) being mono-allergic and only 4 (7.8%) having concomitant allergies to avocado and 2 (3.9%) to banana. In contrast to western studies, latex and pollen allergies were not observed in our patients living in a birch-free tropical climate, thus explaining the low rates of PFAS-associated isolated oro-pharyngeal symptoms. Co-existing food allergies was noted in 68.6% of subjects with more than half (52.9%) having allergies to peanut and/or tree nuts. Kiwi seeds carrying Act d 12 and Act d 13, a major and a minor allergen respectively, have been implicated in the cross-reactivity with peanut and tree nuts because of the similarity in seed storage proteins [7]. The rate of anaphylaxis in our population is similar to other reported studies [8, 9]. Literature suggests that severe kiwi allergy is more likely in young children [6], and patients without pollen allergy [8]. In our study, comparison between anaphylaxis and non-anaphylaxis patients did not reveal statistically significant associations with age, concomitant atopy, and PPT wheal size (results not shown). The natural history of kiwi allergy has not been well-described. Over a median follow-up duration of 2.6 years (range 0–11.7 years), 23 kiwi-allergic patients (45.1%) had interval PPT testing (range 1–5 PPTs), and 4 (7.8%) patients subsequently demonstrated negative PPT results. Although home introduction of kiwi was recommended in the four, no further details on clinical tolerance is available. There are several limitations in this study, aside from it being a retrospective analysis. None of the patients had the diagnosis confirmed with an oral food challenge, component resolved diagnostic was not available, and many patients were lost to follow up. In conclusion, this study represents the largest cohort of kiwi-allergic children in Asia demonstrating several unique aspects: predominance of mono-allergy to kiwi without concomitant pollen or fruit allergies, co-existing allergies to peanut and/or tree nuts and the predominance of cutaneous rather than oral pharyngeal symptoms. Although PPT with raw kiwi has been shown to be useful in diagnosing kiwi allergy, future studies including component resolved diagnostics would be valuable in better defining the predominant allergens in our population. S.M. and K.W.C. designed the study, contributed to data collection and drafted the initial manuscript. S.H.G. performed the statistical analysis. S.H.G., W.L. and K.W.C. interpreted the results and critically revised the manuscript. All authors read and approved the final manuscript. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.