sCD14 and intestinal fatty acid binding protein are elevated in the serum of patients with idiopathic anaphylaxis

Abstract

Background:

Intestinal epithelial integrity compromise has been identified in gastrointestinal (GI), atopic, and autoimmune diseases.

Objective:

Episodes of idiopathic anaphylaxis (IA) are often accompanied by GI manifestations. We therefore sought to determine whether surrogate markers of GI permeability were aberrant in this patient population.

Methods:

Serum concentrations of zonulin, intestinal fatty acid binding protein (I-FABP), and soluble CD14 (sCD14) measured in 54 patients with IA were compared to concentrations in healthy controls (HCs); and correlated with clinical and laboratory parameters.

Results:

I-FABP was elevated in sera of patients with IA compared to HCs (median 1378.0 pg/mL vs 479.0 pg/mL respectively, p < 0.001). sCD14 was also elevated compared to HCs (median 2017.0 ng/mL and 1189.0 ng/mL respectively, p <0.001), whereas zonulin was comparable between patients with IA and HCs (median 49.6 ng/mL vs 52.4 ng/mL respectively, p = 0.40). I-FABP was elevated in patients with IA who experienced vomiting and/or diarrhea compared to patients with IA who did not (p = 0.0091).

Conclusion:

I-FABP and sCD14 are elevated in the serum of patients with IA. Elevations in these biomarkers of IA provides evidence that increased gastrointestinal permeability, as is observed in other allergic conditions such as food allergy, is a common finding in those with IA and offers possible insight into the pathogenesis of this disease.

Introduction:

Idiopathic anaphylaxis (IA) is defined by a periodic acute onset of symptoms usually involving the skin and mucosa, in addition to one or more of the following: respiratory compromise, reduced blood pressure, and associated end-organ dysfunction in the absence of known provocations such as food, medications, hormones, insect venom, and physical factors (i.e. exercise)(1, 2). By definition, the pathogenesis of anaphylactic episodes in IA is not well understood.

Based on a questionnaire in 1995 to graduates of the Northwestern University Allergy-Immunology Fellowship training program with extrapolation of the cases of IA reported to the approximately 4000 allergists then in the United States, the estimated number of cases in the United States was estimated to be between 20,592 and 47,024(2, 3). Based on two independent, nationwide, cross-sectional random-digit-dial landline telephone surveys conducted between July and November 2011, the prevalence of anaphylaxis in the general population was estimated to be at least 1.6% and probably higher, and within this group 39–59% were estimated to be idiopathic (4). A greater incidence of anaphylaxis in women than in men has been described in a review of 601 cases of anaphylaxis where females comprised 62% of cases (5)and a history of atopy was estimated to be as high as 58–80% in patients with IA (2, 6). Mastocytosis, which is a disease characterized by the expansion and accumulation of neoplastic mast cells in one or more organ systems, has notably been found to be an important risk factor for anaphylaxis, and therefore clonal mast cell disease should be considered in a differential diagnosis of IA (2).

Persistent gastrointestinal symptoms are included in the diagnostic criteria for IA (1). The pathophysiology of gastrointestinal manifestations during anaphylaxis is not understood, although are usually attributed to the release of mast cell mediators. Of note, a number of diseases including mastocytosis (7), celiac disease (8–10)and Crohn’s disease (11), which have overlapping gastrointestinal symptoms with those of anaphylaxis, have noted possible compromise in GI permeability as measured by elevations in surrogate serum markers of gastrointestinal permeability. Atopic inflammatory diseases including food allergy (12), asthma (12–15), allergic rhinitis (16) and atopic dermatitis (17) have also been shown to have perturbations in microbial translocation markers (MTMs), supporting the hypothesis that disturbance of the epithelial barrier by a variety of insults may play a role in many chronic noncommunicable diseases (18). Considering the overlap in symptoms of IA with diseases which have a compromise in gut permeability, we sought to explore if this is a feature of IA through the measurements of serum microbial translocation markers. We targeted the measurement of serum zonulin, intestinal fatty acid binding protein (I-FABP/FABP2), and soluble CD14 (sCD14) which have been identified as reliable markers of gut permeability and epithelial integrity (19).

Zonulin is a 47 kDa protein, a precursor to the hemoglobin scavenger haptoglobin 2, produced by intestinal and liver cells, and functions by reversibly regulating intestinal tight junctions and therefore intestinal permeability (20, 21). Intestinal fatty acid binding protein (I-FABP/FABP2) is part of the fatty acid binding protein family, consisting of small 14–15 kDa cytosolic proteins that play an important role in transportation and metabolism of long-chain fatty acids (22, 23). I-FABP is specifically and abundantly present in mature epithelial cells of the mucosal layer of the small intestinal tissue and released into circulation after small intestinal mucosal tissue injury, making it a non-invasive marker for evaluating gastrointestinal wall integrity loss and inflammation (22). Lastly, sCD14 is one of two forms of CD14 found in the human body, produced by shedding of the membrane associated glycosylphosphatidylinositol tail found in mCD14 under catalysis of protease or phospholipase upon stimulation of CD14; or secretion via intracellular vesicles (24). It functions with LPS binding protein, as a cofactor to mediate LPS recognition and response by Toll-like receptor 4, found on several immune cells, and when myeloid cells are stimulated by LPS in vivo, they secrete sCD14 and plasma concentrations of sCD14 are increased (PMID 17115046) (25). Moreover, recent work suggests that analysis of plasma concentrations of sCD14 in conjunction with biomarkers associated with damage to the GI tract was a reasonable surrogate for analysis of concentrations of translocation of microbial products (PMID 32426577).

In a cohort of 54 patients with IA, we thus measured serum concentrations of zonulin, I-FABP, and sCD14 to determine if their concentrations differed from healthy controls and correlated with any symptoms or laboratory findings. Results were compared to findings in other disease states including mastocytosis.

Methods:

Study Patients

Following informed consent on the National Institutes of Health (NIH) protocol 08-I-0184, (Clinical Trial Number: NCT00719719), 54 patients with IA, aged 21–78 years (median 54), were evaluated clinically which included a bone marrow biopsy and aspirate to rule out a clonal mast cell disorder. Fourteen HCs aged 32–74 (median 58) were evaluated under NIH protocol 09-I-0049 (Clinical Trial Number: NCT00806364) following informed consent. Both protocols were approved by the Institutional Review Board of the National Institute of Allergy and Infectious Diseases. All patients self-reported their sex and race and ethnicity during the clinical evaluation. Both groups provided blood samples for analysis. Patients referred with IA were required to have had an evaluation by a primary care provider, or in an emergency room, or hospital, to determine if symptoms were consistent with anaphylaxis based on established criteria, which include a history of involvement of the skin and/or mucosal tissue (generalized hives, pruritus or flushing, swollen lips-tongue-uvula), respiratory compromise (dyspnea, wheeze-bronchospasm, stridor, reduced peak expiratory flow, hypoxemia), reduced blood pressure or associated symptoms of end-organ dysfunction (collapse, syncope, incontinence), and/or persistent gastrointestinal symptoms (crampy abdominal pain, vomiting) (1). Symptoms reported in this study were directly associated with anaphylaxis and were not chronic in nature. Outside records of patients with IA were reviewed independently of the referring physician by the PI and the research team. Objective findings were documented in their medical records. In the analysis of severity of the reaction; organ systems involvement was divided into four groups: dermatological, respiratory, cardiovascular, and gastrointestinal. The evaluation of all patients with IA at NIH did not take place in proximity to a known anaphylactic event and included a detailed history, physical examination, laboratory evaluation, and the determination at time of evaluation if there were any current symptoms consistent with anaphylaxis (negative in all patients). Although some patients had a history of food, drug, or venom allergy, their manifestations did not include anaphylaxis.

Measurement of Microbial Translocation and Epithelial Damage Biomarkers

Blood samples were taken from each of the 54 patients with IA at baseline and analyzed, with a median time of 48 days (IQR 32–93 days) between the time the blood sample was obtained and the most recent event. I-FABP was measured per Hycult Biotech’s (Uden, the Netherlands) ELISA kit instructions (detection range: 47–3000 pg/mL); sCD14 plasma levels were evaluated per R&D Systems’ ELISA (Minneapolis, Minnesota, United States) kit instructions (detection range: 250–16000 pg/mL); and Zonulin was measured per IDK ELISA (Bensheim, Germany) kit’s instructions (detection range: 0.25 to 16 ng/mL).

Statistical Analysis

All statistical analyses were performed using Prism v9.3.1 (GraphPad Software). Mann-Whiney tests were used for direct comparison of MTMs between patients with IA vs HCs; patients with IA who experienced a given symptom vs patients with IA who did not; and patients who experienced an anaphylactic event with symptoms affecting two, three or four organ systems. MTMs were also correlated with laboratory markers using linear regression.

Results:

Participant characteristics

The ages of 54 patients who met the inclusion criteria for the diagnosis of IA (median age of 54) and the ages of the HCs (median age of 58) were not found to be significantly different (p = 0.307) (Figure E1A). Regarding sex, 72.2% (39) were female, and the greater proportion of females to males appears consistent with the observation that IA is more common in females than in males (26)(Table 1). No differences were found in measured MTMs when comparing females to males (Figure E1B–1D), and therefore matching the control group for a similar distribution of sex was not an essential consideration. In the patient population, 5.6% (3) were Asian, 1.9% (1) were Multiracial, and 92.6% (50) were White. Both the patients with IA and the HCs included in the study were predominantly White, thus the HC group was evaluated to be comparable to the group of patients with IA. The source of the classification for both sex and race and ethnicity was obtained through self-report during the clinical evaluation.

Table 1.

Subject Demographics

	IA	Healthy Controls
Age (n)	54	14
Median (IQR) in years	54 (42–63)	58 (44–68)
Sex
Female	72.2% (39)	57.1% (8)
Male	27.8% (15)	42.9% (6)
Race and Ethnicity
African American		28.6% (4)
Asian	5.6% (3)	7.1% (1)
Multiracial	1.9% (1)
White	92.6% (50)	64.2% (9)

Of the patients diagnosed and evaluated with IA: 90.7% of patients experienced dermatological symptoms of flushing (55.6%), hives (48.1%), and/or angioedema (48.1%); 72.2% of patients with IA experienced respiratory manifestations during an episode such as laryngospasm (48.1%), shortness of breath/wheezing (48.1%), and/or asthma (22.2%); 63.0% of patients with IA reported cardiovascular symptoms including pre-syncope (63.0%) and/or loss of consciousness (16.7%); and 53.7% of patients experienced GI symptoms of vomiting and/or diarrhea (Table 2). These symptoms were documented as presenting only in association with an acute event.

Table 2.

Table 2. Symptoms

Symptoms during Anaphylaxis	N (% of 54 total)
Dermatology	49 (90.7%)
Flushing	30 (55.6%)
Hives	26 (48.1%)
Angioedema	26 (48.1%)
Respiratory	39 (72.2%)
Laryngospasm	26 (48.1%)
Shortness of breath/wheezing	26 (48.1%)
Asthma	12 (22.2%)
Cardio	34 (63.0%)
Pre-syncope*	34 (63.0%)
Loss of Consciousness	9 (16.7%)
GI (vomiting and diarrhea) Δ	29 (53.7%)

^*Pre-syncope as defined by feeling faint (i.e. lightheaded, dizzy, sweating)

^ΔSevere abdominal pain not reported as an acute isolated symptom

MTM profile of IA

I-FABP serum measurement showed significant elevations in patients with IA when comparing patients with IA (median 1378.0 pg/mL) vs HCs (median 479.0 pg/mL) (p<0.001). sCD14 concentrations were also found to be significantly elevated in those with IA (median 2017.0 ng/mL) when compared to HCs (median 1189.0 ng/mL) (p<0.001). Serum zonulin concentrations in patients with IA did not significantly differ from that of HCs (median 49.6 ng/mL vs 52.4 ng/mL respectively) (Figure 1A–C). Abnormal concentrations of I-FABP and sCD14 as serum biomarkers of gastrointestinal permeability are thus common in those with IA and measurable between episodes of anaphylaxis.

Figure 1. MTMs in Idiopathic Anaphylaxis vs Healthy Controls.

A. I-FABP was significantly elevated in patients with IA (median 1378.0pg/mL) compared to healthy controls (479.0 pg/mL, P <0.001). B. sCD14 was also significantly elevated in patients with IA (median 2017.0 ng/mL) compared to healthy controls (1189.0, P <0.001). C. Median zonulin levels in patients with IA (49.6 ng/mL) were not significantly different compared to healthy controls (52.4 ng/mL).

MTMs were then correlated with standard laboratory tests including CMP, tryptase, hematological analyses (bone marrow, KIT D816V mutation, mast cell immunophenotyping, CBC + diff, and clotting factors analyses), IgE analysis, and hepatitis and HIV screening. While I-FABP, sCD14 and zonulin concentrations positively or negatively correlated with a number of standard laboratory tests, it should be noted that the concentrations of these standard laboratory measurements were within the normal range. Absolute monocyte levels, however, were elevated in our IA cohort over HCs (median of 0.51 k/uL vs median 0.36 k/uL respectively, P < 0.001), but did not correlate with sCD14 nor other MTMs (P = 0.80, R squared = 0.001252) (Figure E2). It has been reported in an animal model that mice depleted of monocytes and macrophages displayed diminished anaphylactic responses (27) suggesting a role for monocytes levels in anaphylaxis, but the relationship between peripheral blood monocyte and sCD14 levels remains unclear. Basal serum tryptase concentrations in patients with IA (median of 4.17 ng/mL) were not significantly different (P = 0.35) than those of healthy controls (median 3.95 ng/mL) (Figure E3). Thus, the abnormalities in gastrointestinal permeability observed in those with IA are not accompanied by obvious abnormalities in standard laboratory tests, making the abnormal serum values of MTMs unique as serum biomarkers of IA.

Symptoms vs MTMs

Serum I-FABP concentrations were found to be significantly elevated in those with the GI symptoms of vomiting and/or diarrhea vs those without vomiting/diarrhea (p = 0.009) (Figure 2 A–B), though the difference in zonulin and sCD14 between those with vomiting/diarrhea vs those without did not reach significance. I-FABP, sCD14 and serum Zonulin, concentrations were not found to be elevated in patients with IA who had experienced symptoms of flushing, hives, angioedema, laryngospasm, shortness of breath/wheezing, or asthma) compared to those with IA without these symptoms (Figures E4–6). Thus, outside of I-FABP concentrations and vomiting and diarrhea, MTM concentrations did not predict clinical correlates.

Figure 2. MTMs in IA Patients with Vomiting/Diarrhea vs Patients Without Vomiting/Diarrhea.

A. I-FABP was found to be significantly elevated in patients with IA who experienced vomiting and/or diarrhea vs those who did not (P = 0.0091). No such difference was found between these two groups when comparing zonulin and sCD14 (2B, 2C respectively).

Of the 54 patients with IA, a subset reported a specific history of food allergy (9, 16.7%), drug allergy (21, 38.9%) and/or venom sensitivity (2, 3.7%) but episodes of anaphylaxis did not correlate with exposure to these allergens and agents. A significant number of patients (40, 74.1%) reported some history of an allergic disease including allergic rhinitis, asthma, urticaria, or atopic dermatitis. This is consistent with other reports of an atopic history in those experiencing unexplained episodes of anaphylaxis (28). When comparing the concentrations of MTMs in patients with and without these histories, no correlations were found. Comorbidities in the patient group reported to be related to elevated concentrations of MTMs, including food allergy (10) (12, 29), IBD (2) (11, 30), IBS (2) (31), arthritis (2) (32), PCOS (2) (33), and sleep apnea (3) (34), were rarely identified, were not associated with elevations in MTM concentrations, and were not found in the control group.

MTMs were also analyzed according to the number of organ systems involved whereupon increased organ system involvement could be indicative of episode severity (Figure 3, A–C). While no comparisons reached statistical significance, some trends were noted. Patients with IA and symptoms across three organ systems had higher I-FABP concentrations than those with symptoms across two organ systems (p = 0.059), and the median for I-FABP concentrations in patients with symptoms across four organ systems was elevated when compared to those with two organ systems involved (p = 0.09). This trend was not observed when comparing the concentrations of zonulin or sCD14 in patients with varying degrees of organ system involvement Thus, in particular, there was suggestive evidence that higher concentrations of I-FABP are observed in those with multiple organ involvement.

Figure 3. MTMs by Organ Systems Affected.

MTMs were evaluated in patients with IA by varying numbers of organ systems involved. A. Elevations in median I-FABP for those with three organ systems involvement vs two approached significance(P = 0.059), and the median for four systems involved was greater compared to three. This trend was not found in sCD14 or zonulin (3B, 3C respectively) comparing the same subsets.

Because patients with mastocytosis may have episodes of anaphylaxis, and mast cells have been implicated in episodes of IA, we next compared MTM concentrations in those with IA with values for MTM in patients with mastocytosis previously reported from our group (7). Mastocytosis is characterized by an expansion of clonal mast cells at tissue sites which including the gastrointestinal (GI) tract. Somewhat like IA, clinical features of mastocytosis include pruritis, flushing, vomiting and/or diarrhea, abdominal pain, and cardiovascular instability, and both IA and mastocytosis are classified as mast cell activation syndromes. We found both an overlap and a divergence in the MTM profiles associated with IA vs mastocytosis (7)(Table 3). Elevations in both I-FABP and sCD14 are thus observed in both conditions. However, the MTM profile of IA diverges from that of mastocytosis, with a lack of zonulin elevation in those with IA. One conclusion appears to be that various mast cell disorders carry specific profiles of MTMs.

Table 3.

MTMs in IA compared to Mastocytosis and HCs

Variable	IA	Mastocytosis*	Healthy Control	p-Values
I-FABP (n)	52	67	10
Median (IQR) in pg/mL	1378.0 (874.8–2263.0)	629.0 (454.1–929.0)	479.0 (226.1–639.3)	<0.0001 IA-Healthy Control 0.0262 Masto-Healthy Control <0.0001 IA-Masto
sCD14 (n)	54	67	10
Median (IQR) in ng/mL	2017.0 (1800.0–2279.0)	2180.0 (1927.0–2709.0)	1189.0 (1025.0–1355.0)	<0.0001 IA-Healthy Control <0.0001 Masto-Healthy Control 0.0150 IA-Masto
Zonulin (n)	54	67	13
Median (IQR) in ng/mL x20	49.6 (41.9–63.0)	71.0 (63.2–91.0)	52.4 (38.0–54.8)	<0.0001 IA-Masto <0.0001 Masto-Healthy Control

^*Reference (7) describes MTMs data from patients with mastocytosis.

The 67 patients with mastocytosis included in this study had a median age of 57 (IQR 51–66 years old), 69% were female, 94% were White, with 3% of patients included being African American, 1% of patients included being Multiracial, and 1% of patients not specifying in their responses.

This led us to further compare MTMs in IA with concentrations in other diseases (Table 4). I-FABP has been found to be elevated in patients with gastrointestinal disease, such as ulcerative colitis vs HCs (30), and pancolitis vs left-side colitis (30)as well as patients with primary sclerosing cholangitis (35). Similarly, a notable overlap in MTM profiles exist between IA and other atopic diseases including asthma (12–15), allergic rhinitis (16), atopic dermatitis and/or food allergy, food allergy and/or bronchial asthma (12), or atopic dermatitis (17). I-FABP is also reported to be elevated in patients with chronic plaque psoriasis (36). The commonly found elevations in I-FABP and sCD14 concentrations in the GI, atopic, and autoimmune diseases suggest possible overlap in either symptomatology, pathogenesis, or pathophysiology between these diseases and IA.

Table 4.

Comparison of MTMs in Other Diseases

Comparison of Microbial Translocation Markers	I-FABP	sCD14	Zonulin
Mast Cell Diseases
Idiopathic Anaphylaxis	↑	↑	=
Systemic Mastocytosis(7)	↑	↑	↑
Gastrointestinal Diseases
Celiac Disease(8–10)	↑
Inflammatory Bowel Diseas(11, 28)	↑
Primary sclerosing cholangitis(29)	=	↑
Atopic Diseases
Asthma(14, 15)		↑a ↓b	↑
Allergic rhinitis(16)		↑
Atopic dermatitis(12, 17)			↑
Food allergy(12)			↑
Food sensitivity(31)	↑	↑
Autoimmune Disease
Chronic plaque psoriasis(30)	↑

↑ or ↓ indicates significant change in comparison to control or remission

Blank indicates not measured

^aIndicates higher vs healthy control

^bIndicates lower in severe vs mild asthma

Discussion:

Here we report that patients with IA exhibit elevation of the serum MTMs I-FABP and sCD14 between episodes of anaphylaxis and where these markers are reflective of an increase in GI permeability. I-FABP was shown to be significantly elevated in patients with IA, particularly those who experienced the GI symptoms of vomiting and/or diarrhea. While not direct evidence, the correlation of vomiting/diarrhea with the elevation of I-FABP suggests GI symptoms may play an important role in event presentation. Additionally, I-FABP sera concentrations trended higher in those with a greater number of categorized organ systems involvement.

This study interestingly documents both an overlap and a divergence in the MTM profiles associated with IA vs mastocytosis, an important risk factor for anaphylaxis (7) (Table 3). Somewhat like IA, clinical features of mastocytosis include pruritis, flushing, vomiting and/or diarrhea, abdominal pain, and cardiovascular instability, and both IA and mastocytosis are classified as mast cell activation syndromes. The finding that the MTM profile for IA overlaps with mastocytosis, as in the case of elevation in both I-FABP as well as sCD14 observed in both conditions at baseline, as well as the finding that the MTM profile of IA diverges from that of mastocytosis, as in the case of the lack of zonulin elevation observed in IA, the greater I-FABP elevation observed in those with IA compared to mastocytosis, and the slightly lower elevation of sCD14 when compared to mastocytosis, is provocative and provides the groundwork for further investigation as to whether various mast cell disorders carry specific profiles of MTMs (Table 4).

The MTM profile of IA at baseline also has notable overlap with other GI, atopic, and autoimmune diseases (Table 4). I-FABP has been commonly studied in GI diseases, such as celiac disease (CD) and was found to be elevated when examining pediatric patients with CD vs patients with abdominal pain (10), pediatric patients with CD vs pediatric patients with CD after 6 months on a gluten free diet (10), and patients with CD vs HCs (8). This I-FABP elevation in CD was reinforced in a separate study examining CD in both pediatric patients and adults when compared to healthy pediatric and adult patients as well as when compared to patients with irritable bowel disease (IBD) (9). I-FABP was also shown to be elevated in patients with CD on day 14 of a gluten challenge when compared to the same patients at baseline (8), and patients with CD who adhered to a gluten free diet were observed to have lower I-FABP when compared to the same patients at baseline (9, 10). While pediatric and adult patients with IBD were not shown to have significantly different concentrations of I-FABP when compared to pediatric and adult HC patients (9), I-FABP was found to be elevated when comparing those with Crohn’s disease to HCs, those with active Crohn’s disease compared to the same patients in remission (11), ulcerative colitis vs HCs, and pancolitis vs left-side colitis (30). Comparing MTMs of patients with primary sclerosing cholangitis vs HCs, although I-FABP was found in comparable concentrations between both groups, sCD14 concentrations were found to be elevated in the disease group (35).

Notable overlap in MTM profiles exist between IA and other atopic diseases. sCD14 measured after an acute asthma attack was shown to be elevated when compared to those in recovery (14), although lower in those with severe asthma vs mild asthma (15), while zonulin was found to be elevated in patients with asthma vs HCs (13). sCD14 was also found to be elevated in those with allergic rhinitis vs HCs. Pediatric patients with atopic dermatitis and/or food allergy, food allergy and/or bronchial asthma, or atopic dermatitis were also found to have elevated zonulin concentrations when compared to HCs (12, 17). Patients with wheat sensitivity without celiac disease or wheat allergy (NCWS) also showed elevated I-FABP and sCD14 concentrations when compared to HCs as well as to these same patients after 6 months of dietary restrictions that prohibited wheat, rye, and barley (29). I-FABP was found to be elevated in patients with chronic plaque psoriasis when compared to HCs (36), again providing interesting overlap with the MTM profile in IA. The commonly found elevation in I-FABP and sCD14 concentrations in the GI, atopic, and autoimmune diseases in which MTMs were studied suggest possible overlap in either symptomatology, pathogenesis, or pathophysiology between these diseases and IA.

Higher concentrations of serum indicators of gastrointestinal permeability at baseline in patients with IA may also contribute to the growing body of literature supporting the epithelial barrier hypothesis, which states that disturbance of the epithelial barrier by a variety of elements contributes to microbial dysbiosis and tissue inflammation, partially through the allowance of foreign substances to cross from the gastrointestinal tract into subepithelial tissue (18). A recent report of food allergy and anaphylaxis found that, using a C3H/HeJ mouse model, aberrant gastrointestinal permeability may foster translocation of allergens into circulation, which caused anaphylaxis to ingested food allergens in this strain. Acute exercise has been reported to increase intestinal permeability (37, 38) which has been supported by the identification of increases concentrations of serum peanut protein Ara 6 after consumption in combination with exercise (39). Our findings indicate that patients with IA have increased MTMs which suggest a disturbance of the epithelial barrier. The demonstration of increased barrier permeability needs further exploration as the question is whether it is the result of an underlying epithelial barrier dysfunction that facilitates entry of yet to be identified allergens or whether episodes of anaphylaxis promote epithelial barrier dysfunction.

IA is an exclusionary diagnosis, often requiring a thorough history as well as a battery of diagnostic tests including skin-prick testing, tests for specific-IgE, component resolved diagnostics, and in some cases allergen challenge tests. If an anaphylactic event is suspected in a post-event evaluation, these findings suggest blood analysis of MTM concentrations, particularly I-FABP, could be used as supportive biomarkers in the diagnosis of IA.

In summary, the results of this study on the MTMs profile of patients with IA compared to HCs demonstrates that I-FABP and sCD14 are found to be elevated in IA whereas zonulin remains unchanged, I-FABP concentrations are significantly higher in patients with IA who experienced symptoms of vomiting and/or diarrhea, and that elevated concentrations of I-FABP are more likely to be found in patients experiencing a wider range of symptoms across a greater number of organ systems. MTM elevations in patients with IA provide a demonstrable laboratory abnormality that may provide insight into the pathogenesis of IA including reinforcing a possible allergic etiology, as well as possible epithelial barrier dysfunction due to unknown causes. Also unknown is whether such markers vary over time, particularly as episodes of anaphylaxis in some patients become less frequent.

Highlights Box:

1. What is already known about this topic? – Intestinal epithelial integrity compromise has been identified in gastrointestinal (GI), atopic, and autoimmune diseases.

2. What does this article add to our knowledge? – This study demonstrates that markers of gastrointestinal permeability are increased in patients with idiopathic anaphylaxis, adding to the growing body of literature suggesting epithelial barrier disruption is implicated in allergic, atopic, and/or inflammatory diseases.

3. How does this study impact current management guidelines? – Measurements of serum microbial translocation markers may provide evidence toward a diagnosis in the evaluation of patients with suspected idiopathic anaphylaxis.

Abbreviations:

IA

idiopathic anaphylaxis

GI

gastrointestinal

sCD14

soluble CD14

I-FABP

intestinal fatty acid binding protein

MTM

microbial translocation marker

HC

healthy controls

CD

celiac disease

NIH

National Institutes of Health