The Impact of Food Histamine Intake on Asthma Activity: A Pilot Study

Asthma is a complex chronic inflammatory disorder. Diet’s impact on asthma symptoms is controversial. The objective of this pilot crossover, randomized, two-period study was to examine the effect of dietary histamine intake on asthma symptoms in twenty-one children with mild intermittent asthma. Children were randomly assigned to either a high- or low-histamine diet, based on the Mediterranean pattern, for 4 weeks. After a 2-week washout period, patients crossed to the alternative diet for 4 additional weeks. Asthma symptoms were assessed at baseline and after the completion of each diet period by a clinician. Daily symptoms and peak flow were recorded throughout the intervention. Adherence to the dietary intervention was assessed via analysis of four random 24-h recalls, for each intervention period. Eighteen children completed the study. Significantly higher mean air flow obstruction was recorded and a trend for prolonged and more severe symptoms was observed during the high-histamine period. Diet may have an active and direct impact on asthma symptoms. Food choice is affected and/or may affect symptoms in children with mild asthma. Diet intervention is promising yet challenging, for asthma control.


Introduction
Asthma is a common chronic disorder of the airways, considered nowadays more as an umbrella term to describe a constellation of clinical symptoms, including wheeze, breathlessness, chest tightness, and cough, with various underlying pathophysiologies [1]. Both genetic predisposition and exposure to environmental factors, such as allergens or irritants promoting airway inflammation, are involved [2].
Evidence shows that diet plays a significant role in the development and improvement of asthma control. Mediterranean diet, characterized by high consumption of vegetables, fruits, whole wheat products and fish, has been demonstrated to be beneficial on controlling asthma symptoms [3,4]. In contrast, "Western" diets, containing high amounts of processed food, saturated fat, sugar and salt, have been associated with increased asthma activity [5,6]. Other studies, however, have reached conflicting results on the effect of diet in asthma [7,8].
A number of food ingredients, including histamine, are known to have a pharmacological action and have been suggested to affect the clinical course of asthma, but their exact impact remains Nutrients 2020, 12, 3402 2 of 14 unclear [9,10]. Histamine is contained in food mainly in alcoholic beverages such as red wine or highly processed or spoiled food such as cheese and canned fish, while it can also be found in other food sources after the decarboxylation of the amino acid histidine. Histamine's content in food varies considerably even within the same food product and is highly affected by the product's maturity, storage time and processing [11]. It can be metabolized either by oxidative deamination by diamine oxidase (DAO) or by ring methylation by histamine-N-methyltransferase (HNMT). Notably, the last is regarded as the key enzyme for histamine degradation in the bronchial epithelium [10].
The ingestion of histamine-rich foods results in histamine release and blocks the DAO enzyme that is involved in histamine metabolism [11]. A disequelibrium of accumulated histamine and the incapacity of histamine degradation leads to histamine intolerance [11]. The symptoms of this non-allergic adverse reaction involve several organs, such as skin and gastrointestinal system, but also the cardiovascular and respiratory systems, depending on the plasma histamine levels. Specifically, a slight increase in plasma histamine concentrations (i.e., 1-2 ng/mL) has been shown to exert gastrointestinal symptoms or tachycardia, while a 10-fold higher concentration (i.e., 7-12 ng/mL) can lead to a bronchospasm [11]. As recently reported, food histamine intolerance manifests with the appearance of symptoms such as wheeze and asthma [11] and the proposed mechanism for asthma exacerbations is the reduction of HNMT activity [10].
Undisputedly, cellular histamine is involved in asthma pathophysiology. Apart from the mast cells and basophils, there is evidence that cellular histamine is also produced significantly by some gut microbiome phyla in non-obese asthmatic patients [12][13][14]. Diet has a predominant role in the gut microbiota composition and its metabolic activity, but the precise role of histamine-producing bacteria and of the ingested histamine in the disease course is yet to be discovered.
The present pilot study aims to investigate the potential of controlling dietary histamine intake and whether this can interfere with the development of respiratory symptoms in children with mild asthma.

Patients
Twenty-one children (12 boys), diagnosed and followed up in an outpatient allergy unit of a tertiary pediatric hospital in Athens, with mild intermittent asthma (Global Initiative for Asthma (GINA) Stage 1) [15] and documented forced expiratory volume (FEV1) reversibility, were enrolled in the study.
Exclusion criteria included failure to comply with study procedures, presence of other chronic or acute diseases (except rhinitis, allergic conjunctivitis or atopic dermatitis), systematic treatment for diseases other than asthma, and food allergy history. Children and their parents were informed about the study's protocol and procedures and signed consent was obtained. The parents/caregivers gave their signed informed consent for all the children included in the study before their participation. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Scientifical Council and Bioethics committee of the Children Hospital Panagiotis & Aglaia Kyriakou Scientific (Project identification code: issue 3; 130406).

Study Design
A two-period, two-intervention, randomized crossover design was used that, depending on the histamine content, included two types of diet, a low-histamine (LH) and a high-histamine (HH) diet (Table 1). Participants were randomly assigned following simple randomization procedures (computer-generated random numbers) to start with one out of the two proposed diet patterns for eight weeks and continued with the other one for another eight weeks, with a two-week washout period in between. During this time, the volunteers could consume their typical diet without any restriction. Foods from all groups (dairy, fruits, vegetables, starch, meat, fat) were ranked according to their histamine content into low or high histamine [9,[16][17][18][19]. A food list (Table 1), together with eight model diet plans (1800-2000 Kcal; 50% carbohydrates-15% protein-35% fat of which <10 saturated fat) based on the Mediterranean diet prototype, i.e., with a high content of olive oil, whole wheat products, fish, legumes, fruit and vegetables [3], one for each week of the 8-week intervention period, were prepared (example on Table 2). Considering the high variability on histamine content within the same product depending on maturity, storage and processing, a specialized dietitian encouraged patients and their parents to strictly select food products from the LH or HH food list when they were required to make modifications in the respective daily diet plan. Especially for the LH period they were instructed to select fresh not mature of the respective listed fruit and vegetables, to consume fresh meat and fish and to completely avoid frozen and canned products, as well as ready-prepared or packaged food that contain preservatives. Therefore, food recipes for preparing homemade food like juices, jams, burgers, sauces like tomato sauce, lemonade, bread, salads, cake were provided. Hand measures were used to instruct the food portion size measures [20], but each patient decided his portion size ad libitum to avoid weight changes.
Before randomization (t0), participants were evaluated by a clinician to confirm mild intermittent asthma according to GINA [15] and controlled with low inhaled corticosteroids (fluticasone propionate pMDI 100 µg bid). Basic anthropometric measurements (weight and height) were also recorded.
The best out of three morning and evening Peak Expiratory Flow Rate (PEFR) were measured with a flow meter and recorded on diary cards. A reduction of at least 15% reduction in airflow was considered as an asthma episode [21]. Daily symptom scores [22] were documented on diary cards throughout the study. Upper airway symptoms included blocked/stuffy nose, runny nose, sneezing/itchy nose, itchy/sore/watery eyes, hoarse voice, and sore throat. Lower airway symptoms included cough and wheezing/noisy breathing during the day, the night and during exercise. Each symptom was scored from 0 (no symptom) to 3 (severely troublesome). A sum of at least 3 score units for either the upper or the lower respiratory symptomatology was considered as a "higher score" day. Symptoms were quantitatively associated and compared between the different dietary intervention periods. The recorded symptom scores (upper, lower or total) were evaluated and analyzed.
After the end of the first (t1) and second (t2) intervention period, asthma symptom records were re-assessed on site, and weights were measured. At t1, guidance for the cross-arm diet plan was provided, for patients to proceed to the second intervention stage after the two weeks washout period. During the intervention periods, eight 24 h recalls were recorded randomly by phone during each diet intervention, in order to assess compliance and provide assistance for better diet implementation. Histidine intake, which is metabolized to histamine, together with the overall energy, macronutrients and micronutrients were further determined via the 24-h diet recalls with the Food-Processor Nutrition Analysis Software [23]. A post-hoc analysis of food choices during the intervention period was performed to evaluate preference changes depending on the provided food lists and symptom fluctuation. Adherence to the Mediterranean diet was explored via the KIDMED questionnaire [24]. Participant's flowchart with a description of the patients' inclusion criteria and study's design is shown in Figure 1. implementation. Histidine intake, which is metabolized to histamine, together with the overall energy, macronutrients and micronutrients were further determined via the 24-h diet recalls with the Food-Processor Nutrition Analysis Software [23]. A post-hoc analysis of food choices during the intervention period was performed to evaluate preference changes depending on the provided food lists and symptom fluctuation. Adherence to the Mediterranean diet was explored via the KIDMED questionnaire [24].
Participant's flowchart with a description of the patients' inclusion criteria and study's design is shown in Figure 1.

Statistical Analysis
The distributions of all recorded parameters were assessed with the Shapiro-Wilk test. Descriptive statistics are presented as median (inter-quartile range) for non-normally distributed variables and means ± standard deviation for normally distributed parameters. The Wilcoxon ranksum test and Kruskal-Wallis tests were used to compare continuous variables and the Pearson's X 2 test to compare categorical variables among studied groups. The Wilcoxon matched-pairs rank-sum test was used to compare continuous variables within the same individual between the multiple measurements. All reported p-values are based on 2-sided tests and compared with a significance level of 5%.

Statistical Analysis
The distributions of all recorded parameters were assessed with the Shapiro-Wilk test. Descriptive statistics are presented as median (inter-quartile range) for non-normally distributed variables and means ± standard deviation for normally distributed parameters. The Wilcoxon rank-sum test and Kruskal-Wallis tests were used to compare continuous variables and the Pearson's X 2 test to compare categorical variables among studied groups. The Wilcoxon matched-pairs rank-sum test was used to compare continuous variables within the same individual between the multiple measurements. All reported p-values are based on 2-sided tests and compared with a significance level of 5%. Stata 9.1 for Windows (Stata Corp LP, College Station, TX, USA) was used for all statistical calculations and plots.

Results
Eighteen children (10 boys) (mean age 11.5 ± 3.1 years) completed the study successfully. Mean weight at entry point was 47.3 kg (±11.9), mean height was 1.37 m (±9.3) and according to the respective weight to age and height to age growth charts they were within the normal range. No significant weight changes were observed between the different intervention periods (p = 0.5).
Mean % predicted FEV1 was 92.6 (±8.6) and no loss of asthma control was recorded during the study period. Nevertheless, the mean PEFR during the LH period was 400 L/min (±55.9) and significantly higher (p = 0.0006) than the respective mean PEFR during the HH period 388 L/min (±59). Both upper and lower respiratory symptom scores were lower during the LH period, but the relevant statistical comparisons were not significant. In addition, the median number of symptom-free days had a two-fold increase during the LH diet period as compared to the HH (5.5 days vs. 9.5), but the differences were non-significant (p-value = 0.753) ( Table 3). Table 3. Comparisons between the median number of days with symptoms and median symptom score during the two intervention periods. Patients' dietary habits during the last year before the intervention were characterized by a high consumption of red meat, processed foods, saturated fat, sodium and sugar and low consumption of fruit, vegetables, legumes and fish. In line with this observation, adhesion to the Mediterranean diet was found low with a need to be improved (mean KIDMED score = 4 ± 2).

Type of Diet
Upon intervention, adherence to the diet protocol was assessed via analysis of the 24-h diet recalls, in terms of types of foods consumed (Table 4) as well as macronutrient and micronutrient intake. Histidine intake was found to be higher in the HH diet ( Table 5). As anticipated, the food choices varied among the LH and HH periods. During the HH diet, milk, junk food, juices and vegetables intake was higher at symptom-free days than during the LH diet. At higher symptom score days, the intake of oils and butter, fish, chocolate, egg and junk food were higher in the HH diet than in LH. Fish consumption was generally low but increased during days with higher symptom scores independently of the diet assigned, possibly due to caregivers' belief that is a "healthy" choice to relieve symptoms.
Moreover, symptom fluctuation affected food choices within the same intervention period. In the symptom-free days during the HH diet, children consumed more starchy foods, olive oil, chocolate, bacon, ham or sausages and junk food, than in the days with higher symptoms when they consumed more fish and butter. Respectively, in the LH symptom-free days, individuals chose more starchy products, juices from permitted fruit and permitted ham. On the contrary, at higher symptom days during the LH diet, they consumed more olive oil and fresh fish.
In terms of micronutrients, inadequate intake of vitamin D, potassium, and manganese was noted at all stages of intervention. Marginally lower intake of calcium and folate compared to the Dietary Reference Intakes (DRIs) [25] were noted during LH. Furthermore, although zinc's intake was sufficient at all stages, it was double during the HH stage (Table 5). Changes in food choices are presented in mean (±standard deviation). Comparisons (Wilcoxon matched-pairs rank-sum test) of food choices: 1 during episodes vs. remission periods within the same intervention period; 2 during remission between the two intervention periods; 3 during episodes between the two intervention periods. Deficient intake in specific nutrients is marked in bold letters. ** statistically significant difference (p = 0.05) for specific nutrient's intake among the two types of diet. NA: non-applicable.

Discussion
Childhood asthma is a complex disease and evidence indicate that diet plays a role on its clinical outcome. This randomized crossover dietary intervention aimed to evaluate the effects of dietary histamine in children with mild asthma. Our results clearly show a significantly higher airflow obstruction during the high-histamine period, whereas a trend towards less symptoms and more symptom-free days during the low histamine intake diet periods was observed.
Histamine is a potent inflammatory mediator and is associated with deleterious inflammation in asthmatic patients, triggering airway hyper-responsiveness and remodelling. The microbial community in the intestine is shaped by the diet and supplies the host with a number of metabolites, including short-chain fatty acids and histamine. Histamine-secreting bacteria are found at a higher frequency in faecal samples from asthmatic patients. The possible contribution to their atopic phenotype is a matter of further investigation, through larger-scale studies on the microbial diversity of asthmatics, especially because it is impossible to distinguish the microbial-derived histamine from the human-produced form [14].
Moreover, food histamine has been demonstrated to be a trigger for worsening asthma symptoms [10]. In line with this, we found higher intake of histamine in the form of histidine in the HH period to be associated with higher airflow obstruction and symptom exacerbation. Histamine intolerance in asthmatics could lead to dyspnea or dysphonia provoked after high histamine ingestion [10]. Notably, histidine is not the only source of histamine from food but is also an autochthonous ingredient in some foods produced by different cooking preparations [17]. Therefore, this finding can be partly accepted as a measure of compliance and as a separate notion.
As the role of diet on asthma symptoms is complex, dietary plans implemented herein were designed under Mediterranean-style eating pattern. This was decided in order to control at most the overall nutrient intake and for the benefit of the patients of the study since it is generally accepted that the specific diet pattern reduces inflammation [7]. Indeed, the analysis of the diet-recalls revealed a desirable high consumption of olive oil (MUFA), as the main source of fat at all stages of the intervention.
The anti-inflammatory n-3 PUFA [26][27][28][29] intake though, originating from fish oil, was generally low. Interestingly, when children in both intervention phases developed symptoms, they increased fish consumption. It can be assumed that caregivers considered fish as a "healthy" choice to relieve symptoms. Similarly, our cases presented a deficient vitamin D intake at all times, which has been previously correlated to pulmonary dysfunction in childhood [30], but consumed more butter during the symptomatic periods, which is an external source of vitamin D. Adequate vitamin D intake has been suggested to reduce the number of exacerbations, the requirement of steroids and emergency visits in asthmatic school-aged children. Although no safe conclusions can be produced for the role of vitamin D in this intervention, further evaluations are needed about its beneficial role in asthma, as results so far are unclear [31,32].
Saturated fat intake, originating from milk, egg, fried food, and meat products, was generally high at all intervention stages. Past habits of our participants were closer to the Western diet, with a high intake of SF, refined grains and sugar. Patients were advised at the beginning of the intervention to follow the provided model diets but had the liberty to make changes with foods from the respective provided food list, according to their personal preferences and established dietary habits. As such, they often chose processed foods with higher saturated fat and sugar content. Nevertheless, and although saturated fat is associated with the induction of proinflammatory mediators [33], SF intake was equal in both intervention stages, so no further estimation of its influence can be made.
On the other hand, during the LH period, beta-carotene intake was superior. Beta-carotene is an important antioxidant that has been directly involved in asthma pathogenesis [34]. This positive effect could have added to the shorter and milder periods of symptoms encountered during the low-histamine diet. During the same intervention period, zinc intake was recorded within the recommended DRI's but lower in comparison to the HH period. Zinc has been previously suggested to increase antioxidant function and reduce asthma exacerbations [35]. Nevertheless, these data cannot be better evaluated in the current study's context.
As elucidated from our results, a diet intervention aiming to control a single nutrient's intake is really challenging unless, as elsewhere shown, this is based on a food supplement [36] or ready-prepared meals to control overall intake [37,38]. Both these strategies, though, do not reflect the ability of patients to adapt to the proposed changes of eating habits, nor provide the long-term effect of dietary changes [39]. Irrespective of this, such change is generally difficult to adopt and maintain unless proper therapeutic techniques from medical nutrition and behavioral therapy are incorporated [40].
The present study suggests that food histamine has a significant impact on asthma outcome in children who show a trend for a more "healthy diet" during symptom deterioration periods. Our findings show a significance in mean airflow obstruction and a trend for more symptoms with higher duration during the high-histamine period. As the last did not reach significance, probably due to the small sample of participants, the short intervention period and the mild symptomatology of the selected patients, we could aim for future studies with longer intervention periods, a bigger number of participants or participants with more persistent or severe asthma to confirm this effect.

Conclusions
Eating behaviors change in asthmatics during days with symptoms, as also shown here. Histamine intake, along with other dietary factors, influences asthma. In order for the actual effect on the clinical outcomes to emerge, further exploration with long-term dietary interventions are needed, keeping in mind that a diet intervention is promising yet seriously challenging.