Anaphylaxis to Brazil nut (11 patients) - Pastorello et al. (1998) [153]

Vomiting, diarrhoea, loss of consciousness (1 patient) - Bartolome et al (1997) [18]

Oropharyngeal swelling and itching, facial swelling, laryngeal oedema and bronchospasm with wheezing (4 patients) - Nordlee et al. (1996) [1184]

Facial angioedema, laryngeal oedema, (12 patients) - Arshad et al. (1991) [9]

Angioedema of tongue, laryngeal oedema, urticaria (most severe) - Ewan (1996) [323]

Angioedema of tongue, swelling of eyelids, bronchospasm, generalised urticaria - Hide (1983) [353]

Angioedema, laryngeal oedema, urticaria , diarrhoea, syncope - Gillespie et al (1976) [342]

Commercial nuts extracts or fresh nut extracts.

Nordlee et al. (1996) [1184] made extracts of Brazil nut, transgenic soybean and nontransgenic soybean with 1:10 (w/v) 0.01M phosphate buffered saline, pH 7.4, by rocking overnight at 4°C. These were clarified by centrifugation at 82,000 x g for 1h and filtered through 0.45 µm and 0.2 µm filters into an equal volume of sterile glycerol and stored at 4°C.

Clark and Ewan (2003) [615] classified the wheals as grades 1-5 with SPT 0-2 mm, ≥2 mm, ≥3 mm, ≥ 4 mm etc.

Borja et al. (1999) [862] used prick by prick (PBP) with Brazil nut with a 1:10 w/v extract. A wheal 5 mm greater than saline control was considered positive.

Nordlee et al. (1996) [1184] used 1.8 mg/ml histamine as positive control and saline diluent as negative control. Tests started with 1:1,000,000 dilutions and continued with less dilute extracts until a 3mm diameter wheal was observed in the patients sensitised to Brazil nut.

Clark and Ewan (2003) [615] reported results for 1000 patients in a study of patients allergic to peanut or a tree nut (or several nuts).
Bartolomé et al. (1997) [18] and Borja et al (1999) [862] tested 1 patient each.
Ewan (1996) [323] tested 62 patients aged 11 months to 53 years.
Nordlee et al. (1996) [1184] tested 9 Brazil nut allergic patients and 8 controls.

Clark and Ewan (2003) [615] found that 99.5% had SPT ≥2 mm to the nut which caused the main clinical reaction; 98% had SPT ≥3 mm; and 91.6% SPT > 4 mm. For Brazil nut there was an increase in SPT diameter only when severe reactions (grade 5; median 14 mm) were compared to mild- moderate reactions (grade 1-4; median 10 mm; P = 0.0005). 3% of Brazil nut tolerant patients, who were allergic to peanut or another tree nut, had SPT ≥ 9 mm; only 54% had a negative SPT (0-3 mm); 43% had positive SPT in the range 3-7 mm.

Murtagh et al. (2003) [604] report a positive skin prick test from 10 patients from whom sera were obtained.

Bartolomé et al. (1997) [18] reported that the patient gave a positive skin prick test response to Brazil nut, kiwi and hazelnut extracts, and was negative to regionally specific aeroallergens and other food extracts.

Ewan (1996) [323] reported that 18 patients reacted positively to Brazil nut.

Nordlee et al. (1996) [1184] reported that 9 patients gave a positive skin prick test to Brazil nut extract. 3 of these patients were tested with transgenic soybean extract and gave a positive skin prick test but not to non-transgenic soybean extract while 3 control subjects did not react.

Pastorello et al. (1998) [153] ground the pulp of the nut in a mixer and then defatted it with several passages in acetone followed by paper filtration. The powdered, defatted nut pulp was dried overnight and suspended 1:10 (w/v) in 0.1M phosphate buffered saline, pH 7.4, for 2 hours at 4° to 8°C. The extract was centrifuged, filtered, and stored at –20°C.

Bartolome et al. (1997) [18] ground, defatted, and extracted nuts in Coca's solution by stirring. The extract was centrifuged and proteins precipitated using acetone and lyophilised after sterile filtration.

Nordlee et al. (1996) [1184] extracted Brazil nuts and soybeans with phosphate buffered saline containing 0.02% sodium azide, which was also used as diluent.

Arshad et al. (1991) [9] mixed 0.5g nut powder with 10ml of 0.125M sodium bicarbonate and rotated for 2 hours at 4°C. The extract was centrifuged for 15mn at 7000rpm, filtered, desalted on a PD10 column and lyophilised.

Other studies reference the method of Sun et al. (1987) [1025] for the preparation of Brazil nuts extracts.

Pumphrey et al. (1999) [668] studied 731 patients (age 7 months to 65 years, median 6.6 years) with nut allergy who were tested for IgE to peanut, hazelnut and brazil nut, and had specific IgE > 0.35 kU/l to at least one of these nuts (testing was from January 1994 to August 1998). 282 had IgE to one nut, 130 to two nuts, and 319 to all three nuts. When analysed by gender and age quartile, very similar patterns of IgE reactivity were found in all subgroups though significant age trends and age interactions were found for IgE to individual nuts and combinations of nuts

9/10 patients had positive RAST assay results when collected sera were presented with native Brazil nut extracts in vitro (Murtagh et al. 2003[604]).

Specific IgE (UniCAP, Pharmacia(r)) was determined with the following result: IgE for Brazil nut, class 2 (2.37 kU/l) (Borja et al 1999 [862]).

Pastorello et al. (1998) [153] studied sera from 11 patients with clinical symptoms after ingestion of Brazil nut and a positive CAP result to Brazil nut as well as 10 patients with a positive CAP but no symptoms. The mean CAP result for 6 patients with anaphylaxis was 36.41 kU/l. The difference between symptomatic and asymptomatic patients was highly significant.

The patient serum showed a high level of specific IgE by RAST to Brazil nut (> 17.5 PRU/ml, Class 4) (Bartolomé et al. 1997 [18]).

Nordlee et al. (1996) [1184] reported positive RAST with sera from 9 patients with Brazil nut protein. This could be inhibited with transgenic soybean extract.

Pastorello et al. (1998) [153] separated proteins using 7.5% to 20% gradient SDS-PAGE gel with a discontinuous buffer system. Samples were denatured by heating in 4% SDS and 6% beta-mercaptoethanol at 100°C for 5 minutes.

Borja et al. (1999) [862] separated proteins with 12.5% SDS-PAGE by the method of Laemmli and used a tricine-SDS-PAGE system to study low molecuar mass proteins. Samples were run under both reducing conditions with beta-mercaptoethanol and under non-reducing conditions.

Pastorello et al. (1998) [153] electroblotted proteins to nitrocellulose paper. The blot-nitrocellulose paper was then cut into strips and each was incubated overnight with sera from individual patients under investigation diluted 1:4 (v/v) in blocking solution. To determine the IgE-binding components, the nitrocellulose strips were incubated for 6 hours with ¹²⁵I-labeled anti-human IgE antiserum (Pharmacia Diagnostics AB, Uppsala, Sweden) diluted 1:5. The unoccupied nitrocellulose membrane–binding sites were then blocked by incubation in PBS, pH 7.4 ± 0.2, with 0.1% (v/v) Tween 20 for 30 minutes at 37°C.

Four groups of proteins which bind IgE were identified on immunoblots: <5kD, 15-20kD, 25-40kD, 45-50kD (Arshad et al 1991 [9]).

Nordlee et al. (1996) [1184] reported that sera from 8/9 Brazil nut allergic subjects bound to the 2S protein from Brazil nut and sera from 7/9 patients bound a similar protein from transgenic soybean. Sera from a single patient bound a 42 kDa protein from Brazil nut but not the 2S protein.

Bartolomé et al. (1997) [18] found IgE-binding proteins included the 9 kDa and 4 kDa subunits of the 2S albumin and also the alpha-subunits, 33.5 kDa and 32 kDa, and at least one of the beta-subunits, approximately 21 kDa, of the 12S Brazil nut globulin (also called the 11S globulin).

Pastorello et al. (1998) [153] found that all the Brazil nut allergic patients had specific IgE against a 9 kDa allergen. Other allergenic polypeptides were detected at less than 5 kDa (2 of 11 patients, 18.2%), 14.4 kDa (4 of 11 patients, 36.6%), 22 kDa (4 of 11 patients, 36.6%), 35 kDa (4 of 11 patients, 36.6%), and 48 and 55 kDa (3 of 11 patients, 27.3%). 4 asymtomatic patients showed weak IgE binding at 25 to 58 kDa, which in symptomatic patients represent minor allergens.

With the 12.5% SDS-PAGE gels, Borja et al. (1999) [862] found IgE binding to five bands of 52.2, 41.4, 39, 33.4, and 23.7 kDa under non-reducing conditions. IgE binding under reducing conditions was restricted to two bands of 33.4 and 21.4 kDa. The tricine SDS-PAGE immunoblotting under non-reducing conditions, revealed IgE binding to bands of 19.3, 18.4, and a wide band of 7-12 kDa, probably the 2S albumin. A loss of binding was observed under reducing conditions.

Asero et al. (2002) [863] found that their patient's serum reacted strongly to four distinct Brazil nut proteins at about 18, 25, 33, and 45 kDa, but no reactivity against 2S albumin (10 kDa) was found.

The oral challenge was considered positive if signs or significant symptoms such as urticaria or rhinoconjuctivitis were reported during exposure to the nut, but not the placebo. After termination of the challenge, treatments necessary were administered and the patient observed prior to being accompanied home. Challenges were not performed in cases were there was a history of a serious, systemic or life-threatening reaction or where a good history of reaction had been recorded in the last 5 years.