Anaphylaxis in patients with mastocytosis: a study on history, clinical features and risk factors in 120 patients


Knut Brockow
Department of Dermatology and Allergy Biederstein
Technical University Munich
Biedersteiner Strasse 29
80802 Munich


Background:  Excessive mast cell mediator release may lead to anaphylaxis in patients with mastocytosis. However, the incidence, clinical features and trigger factors have not yet been analyzed.

Methods:  To identify risk factors for anaphylaxis in mastocytosis, we determined cumulative incidence, severity, clinical characteristics, and trigger factors for anaphylaxis in 120 consecutive patients (53 male; 67 female, median age and range 24 years, 1 month to 73 years), and correlated these with disease severity of mastocytosis, skin involvement, basal total serum tryptase, and diaminooxidase concentrations.

Results:  The cumulative incidence of anaphylaxis in patients with mastocytosis was higher in adults (49%; P < 0.01) compared with that in children (9%). Only children with extensive skin involvement had experienced anaphylaxis. In adults, anaphylaxis was correlated to the absence of urticaria pigmentosa lesions (P < 0.03). Reactions occurred more frequently in adults with systemic (56%) when compared with cutaneous mastocytosis (13%; P < 0.02). In adults, 48% of reactions were severe, and 38% resulted in unconsciousness. Major perceived trigger factors for adults were hymenoptera stings (19%), foods (16%), and medication (9%); however, in 26% of reactions, only a combination of different triggers preceded anaphylaxis. Trigger factors remained unidentified in 67% of reactions in children compared with 13% in adults. Patients with anaphylaxis had higher basal tryptase values (60.2 ± 55 ng/ml, P < 0.0001) in comparison with those without (21.2 ± 33 ng/ml), but not diaminooxidase levels.

Conclusion:  Adult patients and children with extensive skin disease with mastocytosis have an increased risk to develop severe anaphylaxis; thus, an emergency set of medication including epinephrine is recommended.

Mastocytosis is a disorder characterized by accumulation of mast cells in the skin and/or internal organs (1). Symptoms result from excessive mast cell mediator release, especially histamine, and may include pruritus and flushing, abdominal pain, diarrhea, dyspnea, tachycardia, or profound hypotension (2–5). Episodes of life-threatening anaphylaxis are a recognized feature of the disease (3, 4, 6–12). Patients with mastocytosis do not differ from the normal population in the prevalence of atopy (13, 14); however, the clinical symptomatology has been reported to be more severe in allergic patients with mastocytosis (8, 15). For example, these patients may suffer from severe anaphylactic shock after hymenoptera stings or during the course of insect venom hyposensitization (7, 8, 10, 16). Fatal anaphylaxis has been described following hymenoptera stings (16, 17), in the perioperative period (18, 19), and of unknown cause in children and adults with mastocytosis (17, 20–23). In some patients with mastocytosis, episodes may be provoked by certain medications, such as nonsteroidal anti-inflammatory drugs (NSAIDs) (24), codeine (25), or narcotics (19, 26). Mastocytosis may also be disguised under the clinical diagnosis of idiopathic anaphylaxis (4, 27, 28).

In the literature, data on the frequency and severity of anaphylactic episodes as well as on trigger factors relevant for patients with mastocytosis are missing. This study aimed to identify risk factors for anaphylaxis in patients with mastocytosis. The history and clinical features of the disease were compared with classification and surrogate markers of mastocytosis, and with diaminooxidase levels. As will be shown, frequency, trigger factors, and individual risk factors for anaphylaxis differ between children and adults with mastocytosis.

Materials and methods


Forty-six children (19 female, 27 male, age: 1 month to 17 years, and median 1 year) and 74 adults (48 female, 26 male, age: 22–73, and median 43 years) consecutively attending the mastocytosis outpatients clinic of the Department of Dermatology and Allergy Biederstein, Technische Universität Munich, between 05/2002 and 12/2005 were analyzed. The diagnosis of mastocytosis and assignment to one of five disease categories according to a World Health Organization classification of systemic mastocytosis (SM) was based on physical and dermatological examination, biopsies from the skin and bone marrow and aspirate findings, presence of a mutation at codon 816 encoding for the growth factor receptor KIT, and total serum tryptase levels >20 ng/ml according to accepted criteria (29). In all children and seven adult patients with serum tryptase values within the normal range (<11.4 ng/ml), cutaneous mastocytosis (CM) was assumed without bone marrow biopsy (30). Adults with mastocytosis were classified into those with CM (n = 13), indolent SM (n = 59), SM with associated hematological nonmast cell lineage disease (n = 1) and aggressive SM (n = 1).


Patients were personally interviewed and all symptoms in the history were recorded on a standardized questionnaire. Anaphylaxis was defined by sudden onset of severe or life-threatening generalized or systemic hypersensitivity reaction in accordance with accepted European Academy of Allergology and Clinical Immunology/World Allergy Organization (WAO) definitions (31). The cumulative incidence of anaphylaxis (attack rate) was calculated as the incidence of anaphylactic reactions since first diagnosis divided by the number of mastocytosis patients.

Severity of anaphylaxis was graded according to Ring and Messmer from grade 1 to grade 4 (32). Patient medication used to treat the anaphylactic reaction was recorded.

Severity of skin involvement

The body surface area affected with skin lesions of CM is given as percentage of total body surface area using the technique employed to assess burns (14). The maximum density of skin involvement was evaluated as percentage of surface of involved skin to the total skin surface in each patient in the area of highest density, as described previously (14).

Serum total IgE, tryptase, and diaminooxidase concentrations

Blood sample was drawn at an anaphylaxis-free interval. Baseline total serum immunoglobulin E (IgE) and tryptase were determined in a fluorometric enzyme immunoassay using the Pharmacia UniCAP system (Phadia, Freiburg, Germany). Diaminooxidase, the main extracellular histamine degrading enzyme, was determined using a radioimmunoassay (Immundiagnostics, Bensheim, Germany). Briefly, serum samples were collected, centrifuged for 10 min at 1000 g and stored at −20°C. Diaminooxidase enzymatically degrades the 3H-thymidine-labeled substrate putrescine-dihydrochloride and activity was determined by quantitating the metabolic product pyrolline in a β-counter. The signal detected was directly proportional to diaminooxidase activity and calculated according to a standard curve (33).

Statistical analysis

The Wilcoxon rank sum (Mann–Whitney U) test was used for evaluating statistical significance between comparisons using spss software. Qualitative analyses were performed using the chi-squared test. A P-value of <0.05 was considered statistically significant. Values are given as median and range.


Cumulative incidence of anaphylaxis in mastocytosis

Demographic and clinical characteristics of patients with mastocytosis are shown in Table 1. The parents of four children (age: 3–8 years, median 3.5 years) (9%) and 36 adults (49%, P < 0.0001, Fig. 1A) reported about a total of 75 anaphylactic reactions (4.3% and 5.5% per year disease duration, respectively). Children experienced between one and four reactions with a median of two; adults between one and five reactions with a median of one, respectively. No biphasic anaphylaxis was reported. There was no difference according to gender and no specific age distribution. Atopic diseases in the patients’ history were neither associated with anaphylaxis, nor more common as expected in the general population (Table 1). Adults without systemic disease (CM) had a lower incidence of anaphylaxis compared with those with systemic disease categories (P < 0.02; Fig. 1B).

Table 1.   Demographic and clinical characteristics of patients with mastocytosis
  1. *Atopic eczema, allergic rhinoconjuncitivits, or allergic asthma.

  2. ISM, indolent systemic mastocytosis; ASM, aggressive systemic mastocytosis; SMAHD, systemic mastocytosis with associated hematological non-mast cell lineage disease; CM, cutaneous mastocytosis; IgE, immunoglobulin E.

Number (n)4674
Male (n)2726
Female (n)1948
Age (years)1 (0–17)43 (22–73)
Disease duration (years)1 (0–13.5)10 (0–45)
 CM (n)4613
 ISM (n)059
 ASM (n)01
 SMAHD (n)01
Surrogates for severity
 Extent of skin lesions (%)40.5 (1–100)60 (0–98)
 Density of skin lesions (%)10 (1–100)10 (0–85)
 Tryptase (ng/ml)6 (1–46)25.5 (3–200)
 Atopic diseases* (%)1128
 Total IgE (U/ml)10 (1–601)12 (1–4058)
Figure 1.

 Cumulative incidence of anaphylaxis in patients with mastocytosis. Adults with mastocytosis have a significantly increased relative risk (P < 0.0001) to develop anaphylaxis compared with children (A). Among adults with mastocytosis, anaphylaxis occurs predominantly, but not exclusively, in those with systemic involvement (B).

Elicitors, symptoms, severity, and treatment

Children.  Parents of four children reported acute anaphylactic reactions after food intake (n = 2), vaccination (n = 1), jump into cold water (n = 1), and without identified cause (n = 6). In contrast to adult patients, hymenoptera stings played no part in eliciting anaphylaxis in children with mastocytosis.

Reactions predominantly involved the gastrointestinal tract [abdominal pain (n = 4), vomiting (n = 3), diarrhea (n = 2)] in combination with generalized pruritus (n = 2) and/or rhinoconjunctivitis (n = 1) as well as dyspnea (n = 2) and tachycardia (n = 2). Treatment consisted of hospitalization and corticosteroids (n = 1), antihistamines (n = 2), and in spontaneous recovery (n = 1).

Adults.  The most common elicitors for severe anaphylactic reactions (grade III) in adults with mastocytosis were hymenoptera stings (27%; 80% wasp stings and 20% bee stings) and unidentified causes (20%), whereas foods (24%) and medication (18%) most often led to milder systemic reactions (grade II, Fig. 2A). Most of eliciting foods, particularly alcoholic beverages [red wine (n = 3), champagne, liqueur, brandy and beer (n = 1) each] and fish, were histamine rich rather than representing frequent food allergens and were specific IgE and skin prick test negative in these patients. Culprit medication included NSAIDs (n = 3), contrast media (n = 2), codeine (n = 1), an amoxicillin (n = 1), and a local anaesthetic (n = 1). Of all anaphylactic reactions, 60% were classified as severe (grade III), and 43% suffered from shock with unconsciousness (in eight patients after hymenoptera stings, 1 × food plus alcohol; in eight patients without identified cause). Alcohol, exercise, mosquito stings and orthostatic exercise were other elicitors mentioned. Twenty-six per cent of reactions occurred only after a combination of several factors, especially alcohol and foods (Fig. 2A).

Figure 2.

 Elicitors and symptoms of anaphylaxis in adults with mastocytosis. Hymenoptera stings, food, and medication were the most common reported elicitors, whereas causes for anaphylaxis remained unidentified especially in severe episodes (A). Patients with mastocytosis developed cutaneous, abdominal, respiratory, and cardiovascular symptoms, with a predominance of cardiovascular symptoms, such as dizziness, tachycardia, and hypotension. Closed bars designate number of patients with, open bars without symptoms (B).

Anaphylactic symptoms in adults with mastocytosis were primarily cardiovascular in nature with dizziness/presyncope (44%), tachycardia (47%), and hypotension or shock (50%; Fig. 2B). Other symptoms especially involved the respiratory tract with dyspnea (44%), gastrointestinal tract with nausea (36%) and diarrhea (33%), and the skin with flush (33%). In 19.2% of patients neither tachycardia, nor hypotension, or dyspnea was recalled, but a combination of other symptoms given in Fig. 2B.

In the 32 patients, who could recall their immediate treatment of anaphylaxis, epinephrine was used in only three of 28 adults (11%) and in no children. In addition, seven adults (25%) were hospitalized, 17 adults (61%) received antihistamines, 13 adults (46%) corticosteroids (Table 2). In nonhospitalized patients, 54% of patients were documented by a physician and 46% were not.

Table 2.   Therapy of anaphylaxis in patients with mastocytosis
TherapyChildren (%)Adults (%)
No therapy2536

Identification of risk factors for anaphylaxis

Children.  None of the children with isolated mastocytomas developed anaphylaxis, although localized flushing occurred and was distressing for the parents. Anaphylaxis was reported in one child with diffuse CM and three children with extensive urticaria pigmentosa and palpable maculopapular or infiltrated lesions demonstrated by an extent of skin lesions of >45% (P < 0.01) and density of skin lesions of ≥15% (P < 0.01; Fig. 3A and B). The severity of cutaneous symptoms was also reflected by increased serum tryptase values compared with children without anaphylaxis (P < 0.03; Fig. 3C). In contrast, serum diaminooxidase levels were not different between children with and without anaphylaxis (Fig. 3D, P < 0.09).

Figure 3.

 Correlation of cutaneous involvement, serum tryptase, and diaminooxidase concentration to presence of anaphylaxis in children and adults with mastocytosis. In children, the extent (A, P < 0.01) and density of skin lesions (B, P < 0.01) did correlate with anaphylaxis, but not in adults. Serum tryptase levels (C) were higher in children (P < 0.03) and adults (P < 0.01) with anaphylaxis, but diaminooxidase levels did not correlate with anaphylaxis (D). Open circles designate children without anaphylaxis, closed circles children with anaphylaxis, open squares adults without, and closed squares those with anaphylaxis. The horizontal bar within each data set is the median.

Adults.  Adults with systemic disease had a higher incidence for anaphylaxis (P < 0.02; Fig. 1B), however, particularly in those lacking additional skin involvement (urticaria pigmentosa) (12%, P < 0.03). If skin involvement was present, the intensity was not associated with the presence of anaphylaxis (Fig. 3A and B). However, serum tryptase concentrations were higher in those with anaphylaxis indicating an association with the severity of systemic involvement. No difference was found with regard to serum diaminooxidase levels (Fig. 3D).


Mast cell mediator release may lead to anaphylaxis in patients with mastocytosis. (2–4, 6–10). This first study on the cumulative incidence of anaphylaxis in a group of 120 patients with mastocytosis shows that anaphylaxis is common in patients with this disease (4.3–5.5% persons per year disease duration) (Fig. 1A). Although figures cannot directly be compared, this appears higher than the expected from incidence figures published in the normal population (3.2–21 per 100 000 persons per year) (34, 35).

Mechanisms, clinical features, and course of mastocytosis differ between children and adults (14), as did the incidence of anaphylaxis in this study (Fig. 1A). In childhood, the risk for anaphylactic episodes was limited to children with extensive skin disease, but nonexistent for children with mastocytoma or limited macular lesions. This is in good agreement with the literature, where children with anaphylaxis were described as having clinically severe skin involvement of mastocytosis, although the levels of skin involvement were not given and tryptase concentrations not determined (36, 37). Children with fatal anaphylaxis, described in three case reports, all had suffered from extensive blistering skin disease (20–22). Trigger factors for anaphylaxis in children remained mostly unidentified in this study (idiopathic anaphylaxis), which is consistent with case reports (20–22), but in contrast to anaphylaxis in children without mastocytosis, where reactions are most commonly caused by food allergens (38). Thus, severe blistering episodes on extensive and infiltrated skin lesions packed with mast cells appear to be the strongest intrinsic trigger for anaphylaxis in children with mastocytosis.

In adults with mastocytosis, symptoms generally appear to be more severe in patients with extensive systemic disease (3, 4, 14). The cumulative risk for anaphylaxis was higher (56%) in those with systemic involvement; however, reactions did also occur in those with CM only (15%, Fig. 1B), which has also been reported in the literature (10, 25). In accordance with this finding, those with anaphylaxis had higher serum tryptase values (Fig. 3C), reflecting a higher total systemic mast cell number and pool of effector cells, possibly leading to a higher risk of spontaneous mast cell degranulation. Patients with SM without skin lesions may be particularly prone to develop anaphylaxis, as highlighted by several case reports and series (11, 23). In accordance with this, in this series of patients, the risk of anaphylaxis was higher in patients with absence of cutaneous lesions compared with those presenting cutaneous lesions of urticaria pigmentosa (P < 0.03). As previously described, patients without skin lesions had more severe internal involvement and higher tryptase levels compared with those with skin lesions (14), and seem to carry a worse prognosis (4). However, in patients with skin involvement, a higher extent and density of skin lesions was not characteristic of those with an increased risk for anaphylaxis (Fig. 3A and B).

The major trigger factors of anaphylaxis in this series – such as hymenoptera stings, foods, and medications (Fig. 2A) – did not differ from triggers of anaphylaxis in patients without mastocytosis (39, 40). Anaphylaxis to hymenoptera stings in mastocytosis is common (7, 8, 10, 16). Atopy did not appear to be a major factor, in contrast to anaphylaxis in nonmastocytosis patients (39, 41). Foods have anecdotically been described as elicitors of anaphylaxis in mastocytosis in the literature (20, 22, 26). In this study, they resulted in a high number of generally mild reactions; however, the causal relationship remains unproven and double-blind placebo-controlled food challenges are yet to be performed. As only a subgroup of patients is affected, theoretically, either a histamine overload or decreased capacity of histamine degradation by the main histamine-degrading enzyme, diaminooxidase, may be responsible for histaminemia leading to anaphylactoid reactions (42). However, in this series of patients, we found no evidence for this hypothesis (Fig. 3D). It is noteworthy that 26% of reactions in adults did occur only after a combination of triggers (summation anaphylaxis), as was previously reported in anaphylaxis in nonmastocytosis patients (39). Alcohol, exercise, and aspirin are not only elicitors, but also important co-factors in anaphylaxis, as has been reported for food allergy, possibly through facilitation of increased food absorption (38). In some cases of our series, elicitors remained unidentified. It has been stated that the diagnosis of ‘idiopathic’ anaphylaxis is relative in nature and depends on the intensity of the allergologic procedures involved (43).

Symptoms of anaphylaxis in this series were primarily cardiovascular, and did often, but not always, also affect the gastrointestinal and respiratory tract (Fig. 2B), as has been described (3, 4). Flush and pruritus were common cutaneous symptoms. Urticaria and angioedema, however, reported to be present in >80% of nonmastocytosis patients with anaphylaxis (39, 41, 44), were uncommon in this series as in the literature (3, 4). Anaphylaxis in patients with mastocytosis was severe in most patients and resulted in unconsciousness in 19 of 36 adult patients (53%). These data confirm previous reports on hymenoptera venom anaphylaxis, where patients with mastocytosis or increased tryptase values had significantly more severe reactions compared with those without mastocytosis (8, 15). Severity of anaphylaxis in this study was not higher in patients with aggressive forms of mastocytosis, and did neither differ between patients with (n = 15/17) and without (n = 2/17) the D816V mutation typical for adult patients with mastocytosis, nor between patients under (n = 5/40) and without (n = 35/40) continuous therapy with antihistamines (data not shown).

Treatment with epinephrine is generally recommended for allergic reactions. However, surprisingly, therapy in the cases of this series consisted mainly of administration of antihistamines and/or corticosteroids; epinephrine was only given in 11% of adults (Table 2). The concordance with guidelines generally has been reported to be very low. A study performed by Clark et al. (44) found that of patients presenting at an emergency department with food anaphylaxis, 72% received antihistamines, 48% received corticosteroids and only 16% received epinephrine. In other studies, only 10% and 12% of patients received epinephrine (45, 46), being similar to the results in the present study. Another factor to consider is that the diagnosis of anaphylaxis can vary according to different clinical definitions, being more (47), or less dependent on recognized allergic trigger factors (31). This is especially difficult for patients with mastocytosis, as acute symptoms of anaphylaxis may be considered to some extent a mast cell mediator-related feature of the disease, but may progress to become severe and nonpredictable (5). Patients with anaphylaxis and mastocytosis have an increased risk to develop repeated anaphylactic reactions (50% of patients >1 episode), similar to other 40–76% patients with anaphylaxis without mastocytosis (39, 41, 48), and should be provided with emergency medication. We do now prescribe and teach proper techniques for self-administration of epinephrine in all adults with mastocytosis and parents of children with severe cutaneous disease. Special care has been taken to apply immunotherapy in hymenoptera allergic patients, as described (10), as well as to use antihistamine premedication and to avoid mast cell-releasing drugs in general anesthesia (49).

In conclusion, in this prospective study, children and adults with mastocytosis differed markedly with regard to anaphylaxis. In children, it was strictly limited to those with severe disease and high tryptase levels, whereas in adults, all patients may be affected. Most common reported trigger factors in adults were hymenoptera venoms, foods and medications; however, in children as in adults, several elicitors remained unidentified and also summation anaphylaxis seems to play a role. Fortunately, in this series, patients responded quite well to treatment or recovered spontaneously. Still it seems justified to recommend that children with severe disease and/or previous anaphylaxis and adults with mastocytosis should be provided with an emergency medication.