My 3-year-old daughter recently broke ou

Answer

Your allergist is correct on the science, and the four questions each have well-established answers in the food allergy literature. Who clears which nut, in what order, with which dose, is her call and her clinic's — not ours. Below is the underlying biology in plain language, with the parts that come from our verified source card flagged separately from general medical knowledge.

A note up front: your daughter's reaction pattern (hives + vomiting after one bite, IgE Class 4, Ana o 3 positive) confirms a real, likely persistent cashew allergy, with pistachio almost certainly a true co-allergy. The decision on the *other* nuts — almond, hazelnut, walnut, pecan, macadamia — is a separate question, and that's exactly the slice the OFC is designed to settle. [Verified per （内部报告路径已省略） (file lines cited below).]

---

## 1. Botanical pairs and cross-reactivity

Your framing is essentially right. The clinical reactivity tracks the protein families, and the protein families track the plant families.

**Cashew ↔ Pistachio (Anacardiaceae)** — both store seed 2S albumins (Ana o 3 in cashew; the homologous 2S in pistachio is the IgE-binding component your card is picking up). The two proteins share high sequence homology, so the same IgE clones cross-link both. Clinical co-allergy rates around **90–100%** are consistently reported in cohort studies, and our pack's verified card states: *"开心果过敏几乎总是与腰果过敏同时存在"* [31-坚果.md › 腰果与开心果诊断] and *"248次评估中235次诊断正确，准确率超过95%"* [31-坚果.md › 腰果与开心果诊断, aai-food-allergy]. So pistachio out of the kitchen — full stop.

**Walnut ↔ Pecan (Juglandaceae)** — same plant family, share homologous 2S albumins (Jug r 1, Car i 1) and 7S vicilins (Jug r 2, Car i 2). Cross-reactivity is real but lower and **dose/asymmetric** — some patients react to pecan but not walnut, or have different threshold doses. This is *not* a >95% pair.

**Almond (Rosaceae, *Prunus dulcis*)** — major allergens are Pru du 3 (LTP), Pru du 4 (profilin), Pru du 5 (PR-10). **None of these share meaningful IgE epitopes with cashew 2S albumins.** Botanical cousin is the peach, not the cashew.

**Hazelnut (Betulaceae, *Corylus avellana*)** — has its own 2S albumin (Cor a 14), but the dominant allergen in many populations is Cor a 1 (PR-10, cross-reactive with birch pollen). Different allergen profile from cashew.

**Macadamia, Brazil nut** — distinct storage proteins (7S/11S globulins, 2S albumins that don't share the cashew epitope set). Generally tested independently.

So the blanket-ban instinct treats all six plant families as one — biology treats them as four.

---

## 2. Sensitization vs. true clinical allergy

This is the single most important conceptual point in your questions.

**SPT wheal and serum sIgE measure IgE binding.** They do not measure whether bound IgE will trigger a clinical reaction on ingestion. The cascade from "IgE cross-links on a mast cell" to "hives, vomiting, anaphylaxis" depends on multiple downstream variables (epitope affinity, IgE clonality, mast cell reactivity, allergen dose, food matrix, gut permeability at the moment). That is why:

- A 3 mm SPT wheal or low-positive sIgE is very often **sensitization without clinical reactivity**, especially for a food the child has rarely or never eaten.
- For tree nuts specifically, our pack states explicitly: *"对多种树坚果的检测敏感（例如皮肤点刺试验所示）不一定代表真实的临床过敏"* [31-坚果.md › 腰果与开心果诊断].
- **OFC is the diagnostic gold standard** [31-坚果.md › 腰果与开心果诊断, aai-food-allergy] precisely because it shows the *whole cascade*.

There is a hierarchy of predictive value:

| Test type | What it measures | Predictive power |
|---|---|---|
| SPT wheal size ≥ 8–10 mm or sIgE above a decision point | IgE binding (varies by food, age) | Moderate PPV (~50% or lower for many foods) |
| Component-resolved diagnostics (e.g., Ana o 3, Jug r 1, Cor a 14) | IgE to a *specific* storage protein | Higher PPV when positive; e.g., Ana o 3 is a reliable predictor of cashew **and** pistachio allergy [31-坚果.md › 腰果与开心果诊断, aai-food-allergy] |
| Oral food challenge | Clinical reaction under observation | **Gold standard** |

This is why your allergist is not waving through the "borderline" walnuts and almonds — borderline IgE is exactly the population where the OFC question matters most. A positive test ≠ she will react; a positive test + a clean OFC = she tolerates it.

---

## 3. The blanket-avoidance problem

The instinct to remove everything is protective and understandable — and for confirmed allergens it is correct. The blanket extension to *tolerated* foods is where the science pushes back.

Two real mechanisms:

**(a) Regular exposure supports mucosal tolerance.** The current evidence base (LEAP for peanut, LEAP-On follow-up, EAT trial, and emerging work for egg and other allergens) shows that early and ongoing ingestion of a food the immune system can handle *trains* regulatory T-cell and IgG4 pathways. Removing a tolerated food removes that input — exactly the opposite direction.

**(b) Loss of tolerance is easier than gaining it.** De novo IgE-mediated allergy to a previously tolerated food is a recognized pattern in toddlers and preschoolers; once tolerance is lost, re-inducing it is hard and slow. For *known* allergens, the management goal is zero exposure; for *tolerated* foods, the management goal is **steady, age-appropriate intake**. These two goals conflict, which is why blanket avoidance of all nuts is the wrong frame.

The card captures the principle: *"对一种树坚果过敏不一定意味着对其他树坚果过敏"* [31-坚果.md › 交叉反应与个体化饮食, aai-food-allergy] — meaning the avoidance list should be individualized to what she actually reacts to, not inferred from what she might.

(Caveat: the strongest trial evidence is for peanut. The tree-nut generalization is biologically consistent and clinically accepted, but the all-cause "blanket avoidance causes de novo allergy" claim is hard to pin to a single randomized trial in tree nuts specifically — your allergist is operating on standard-of-care guidance, not a single citation.)

**Practically for your house:** cashew and pistachio *out*. The other nuts stay in the pantry only if your allergist's plan for them is "in active clinical use or under OFC." Mixed-nut jars, trail mixes, and shared nut butters get reassessed — *those* are cross-contact risks for cashew. Single-nut jars (almond butter, walnut pieces) used only in the home kitchen are a different conversation and is the allergist's call to make once each nut's status is settled.

---

## 4. In-office Oral Food Challenge — protocol and safety

**Where:** A medical facility equipped for anaphylaxis — typically a pediatric allergy clinic or hospital day unit with a waiting area, exam rooms, IV access, oxygen, and crash cart. *Home and clinic OFCs are not equivalent.* Our card states: *"所有口服食物激发试验都应在能够识别和治疗过敏反应的医疗机构内进行，并由过敏专科医生监督"* [31-坚果.md › 口服食物激发试验安全, aai-food-allergy].

**The general shape of a graded OFC:**

1. **Pre-challenge screening** — history, recent antihistamine washout (typically 5–7 days for sedating, longer for cetirizine/loratadine in some protocols), no intercurrent illness, baseline vitals, asthma controlled, recent meal/snack considered.
2. **Dosing schedule** — the food is given in escalating doses at fixed intervals (commonly every 15–30 minutes), starting at a tiny fraction of a normal portion (often 1/100 to 1/50) and doubling until a full age-appropriate serving is reached. Total time in clinic is typically **2–4 hours** plus an observation window afterward (often 1–2 hours after the final dose).
3. **The vehicle matters** — for nuts in young children, the food is usually disguised in a vehicle the child tolerates (oatmeal, applesauce, muffin) to avoid skin/oral contact triggering reactions that confuse interpretation. Dose, not just food choice, is the variable.
4. **What is being watched for** — early objective signs (skin, GI, respiratory, cardiovascular). Subjective symptoms alone (tummy "feels weird") without objective signs usually do not stop the challenge; any objective sign does.
5. **Stopping rules** — defined in advance (typically ≥3 contact hives, any systemic symptom, a 20% drop in peak flow, or wheeze/cough). Anaphylaxis criteria = full stop, treat per her action plan with epinephrine, observation, sometimes IV fluids, discharge criteria determined by the team.
6. **Open vs. blinded** — open OFC (you and she know what she's eating) is often a starting point; double-blind placebo-controlled food challenge (DBPCFC) is the formal gold standard for ambiguous cases but is uncommon in routine clinical follow-up of a clearly positive tree-nut panel.

**Selection of which nut to challenge first** is the allergist's strategic question. Typical approach in a child with confirmed cashew + pistachio allergy and a mixed tree-nut panel:

- Start with the nut most likely to be tolerated (often almond or hazelnut if SPT/sIgE is negative or trace) — passes fast, opens the most dietary space.
- Some clinicians challenge walnut before pecan (or vice versa) because co-reactivity means a pass on one may predict the other; a fail on one is informative for both.
- Macadamia and Brazil nut are often deferred because they rarely appear in the panel strongly positive and are uncommon in the diet.

**Tolerance maintenance at home after a passed OFC:**
Once cleared, the child eats the nut regularly — most commonly **2–3 times per week, in a serving size consistent with age-appropriate intake**, prepared in a form that doesn't involve shared nut jars (so cross-contact is controlled). The principle: tolerance is *use it or lose it*, and gaps of weeks-to-months correlate with re-sensitization in young children. The exact regimen (dose, frequency, vehicle) comes written into the allergist's plan and you keep that as a written record.

---

## Tonight, and what to bring to the next visit

**Tonight (assuming she is well, no current symptoms):**

- Cashew and pistachio stay out of the house in all forms (whole, butter, oil, "vegan cashew cheese," pesto, mixed-nut jars, pistachio butter).
- Carry her epinephrine auto-injector per the allergist's written plan. Confirm both devices (one for caregiver, one for school) are in date.
- A borderline walnut/pecan SPT result is **not** a clearance and **not** an automatic ban; do not feed walnut or pecan at home without the allergist's explicit direction. Same for almond and hazelnut.

**What to bring (or have on file) at the next allergist visit:**

- The actual report pages (not just a verbal summary) so the OFC selection and dose can be planned against the numbers.
- The current action plan and device expiry dates.
- A list of nuts and nut products currently in your pantry — your allergist may want to know which single-nut butters, oils, and baking ingredients are present, to plan the home environment after passes.

One unknown that would change the next step, and worth asking only if you don't already have it: **what specific written plan has the allergist given you about the borderline walnut/pecan results at this visit** — some clinics schedule OFC within weeks, some wait 6–12 months and re-test, some want a specific component-resolved diagnostics panel (e.g., Jug r 1, Jug r 4 for walnut; Cor a 14 for hazelnut) before committing to a challenge. That is the question to put on their front desk.

Sources from our pack cited above: （内部报告路径已省略） (sections 腰果与开心果诊断; 交叉反应与个体化饮食; 口服食物激发试验安全; 口服食物激发试验), with underlying `aaaai-food-allergy` and `ascia-food-allergy` source notes already in that card. Biology section on Anacardiaceae/Juglandaceae/Rosaceae/Betulaceae allergen families and standard OFC dosing/observation workflow is general clinical knowledge, not a quote from our pack; your allergist's own plan and your local clinic's protocol are what actually run the day.