Study design

The study was conducted as a prospective randomized, single blinded, comparative, parallel group design.

Population and sample size

Asthmatic children who were registered at an out-patient department (primary care setting) at Sichon and Pakpanang hospital, Nakhon Si Thammarat, Thailand were enrolled over the period from October5, 2021 to May11, 2022. Inclusion criteria were; being an asthmatic child aged 4–15 years who had previously been diagnosed by a physician as having mildly or moderately persistent asthma following the Global Initiative for Asthma guidelines [37], having previously been treated and prescribed ICS within 24 months prior to the study, having complied with the blood test and pulmonary function test requirements, and having completed the 8-week follow-up for period. Exclusion criteria consisted of; having had a prior airway infection inside a 4 week period prior to the study, had had previous acute asthmatic exacerbation within a 3 months period prior to the study, having had chronic diseases such as cardiovascular, pulmonary, neurology, renal, or psychiatric conditions, having consumed high doses of supplemental substances such as lycopene (60 mg/day), beta-carotene (30 mg/day) or ascorbic acid (180 mg/day) within a 3 month period prior to the study [38], having refused to consume tomato with mixed fruits juice consumption, and being lost in the follow-up.

Sample size was calculated in accordance with a previous study [31] which was conducted in an exercise-induced asthma patients by using lycopene diet compared to placebo. The FEV1 predicted percentage, the mean and standard deviation of both groups were used in the formula of our study as follow;

n / group = 2 (z α/2 + z β)2 Ó2/(x1-x2).2 [39]

Defined α = 0.05, β = 0.10,

Z α/2 = Z 0.05/2 = 1.96 (two tails), Zβ = Z 0.10 = 1.28.

x1 = mean FEV1 predicted percentage in the intervention group = 79.

x2 = mean FEV1 predicted percentage in the control group = 67.

n1 = 20, s1 = 20; n2 = 20, s2 = 16.

Ó2 = Pooled variance.

 = [(n1-1) s12 + (n2-1) s22] / n1 + n2—2.

 = [(20–1)(20)2 + (20–1)(16)2] / 20 + 20–2 = 328.

Then n/group = [2(1.96 + 1.28)2(328)] / (79–67).2

 = 47.8, then n /group = 48, the total number of participants was 96.

Due to the Covid-19 pandemic, many potential participants were afraid of contacting the disease during the out-patient visit, so the enrolment rate was limited. Nevertheless, the percentage of participants who met the inclusion criteria and complied with the study requirements was 83.3% (80 from 96). The study was approved by the Human Research Ethics Committee of Walailak University (WUEC-20–362-01) and participants/guardians gave written informed consent/assent before enrolment. This study also adhered to CONSORT guidelines.

Protocol

This was a parallel, randomized, controlled trial with 2 arms: conventional or usual care (control group) and intervention with usual care (experimental group) as shown in Fig. 1. Participants were randomly allocated to one of the two treatment arms, and this was determined by computer-generated random sequence. Initial arm allocation was concealed from the clinical recruitment investigators until each participant had entered the trial, had received a randomization code, was prescribed tomato with mixed juices and was followed-up until the end of the study. After enrolment, participants and guardians were interviewed by using a modified questionnaire from a previous childhood asthma study [40]. Physical examination, asthma control test assessment, pediatric asthma quality of life questionnaire, pulmonary function test were performed respectively. Blood tests for serum lycopene, beta-carotene and ascorbic acid were carried out by trained staff. The assigned research assistance was only one who knew the participant in each group. This staff gave tomato with mixed juices, provided health education and continuously contacted participants via mobile-phone or line network during the 8-week period to the experimental group. In control group, participants were given health education for usual asthma care. At the 2nd visit, all participants/guardians were processed as per the initial visit.

Fig. 1

Consolidated Standards of Reporting Trials (CONSORT) diagram showing the flow of participants in the study

Participants flow

Eighty-six eligible participants were informed. We excluded 5 participants because they refused to participate (n = 3) and did not meet inclusion criteria (n = 2). Then, on the 1st visit, participants were randomly assigned to the control group (n = 32) or to the experimental group (n = 48). The trial was completed by 80 of 81 participants with only one from the control group lost to follow-up as described in Fig. 1.

Questionnaires

The questionnaire was a modified version of one that had been used in a previous childhood asthma study [40]. It included sociodemographic characteristics, history of related allergic conditions, asthmatic clinical data, medication utilization, and a recent physical examination. The interview processes were performed by the principal investigator, a pediatric allergist.

Asthma control test (ACT)

The Thai version of the ACT was modified for this study. We assessed the score by interviewing a participant and/or a guardian. The ACT consisted of 5 items that related to asthmatic clinical symptoms and reliever medication use within the prior 4 weeks. A total score was 25, and a score of 20 or higher indicated that the participants asthma was well controlled. A patient whose score was less than 20 was said to have asthma that was uncontrolled [41, 42].

Pediatric asthma quality of life questionnaire (PAQLQ)

We modified the Thai version of PAQLQ from a previous study [43]. The questionnaire was comprised of 23 close-ended questions. There were 3 categories; (1) activity limitation—5 items, (2) clinical symptoms—10 items, and (3) emotional aspects—8 items. The PAQLQ scores were ranked from 1–7. A score of 1 indicated most bothersome or all-day symptoms, whereas 7 indicated untroublesome symptoms or an asymptomatic case.

Pulmonary function test

Pulmonary function was assessed by using the mobile spirometer (Vyntus™ SPIRO, Vyaire Medical, Inc. Mettawa, IL, USA) in accordance with the American Thoracic Society/European Respiratory Society guidelines [44]. After each participant took a rest for 15 min, they then sat in a seat with an upright posture and placed both feet on the floor. A nasal clip was applied and then each participant took a deep inspiration while their feet were remained in contact with the floor. A mouthpiece was applied to their mouth, and they then made a rapidly and forceful expiration until the end of expiration cycle. This process was repeated at least 3 times and chose the one which followed the standard criteria. Forced expiratory volume in one second (FEV1), forced vital capacity (FVC), percent of FEV1/FVC, and forced expiratory flow between 25 and 75% (FEF25–75%) were used for analysis. Participants were measured with spirometry both at the initial stage and 8 weeks later at the follow-up visit.

Serum antioxidant levels

We measured serum lycopene, beta-carotene and ascorbic acid for all participants both at the first and follow up visit. Participants were not allowed to eat or drink 10–12 h before performing a blood test in the morning. Sterile needle was inserted at the venule vessel at the cubital fossa, and then 3 ml of venous blood was drawn and placed into a test tube. Centrifugal force was used at 3,000 g, and at 5 °C. Then, the centrifuged blood was stored at -20 °C in a container with dry ice before analysis was undertaken using a High Performance Liquid Chromatography system (Agilent Technologies 1260 Infinity II HPLC system, Palo Alto, CA, USA). The system had a G7111A quaternary pump and G7114A UV–Visible variable wavelength detector that was controlled and processed by an Agilent OpenLab ChemStation Workstation, Version C.01.10 (201). In addition, the system was connected to a HP-compatible computer, which reported the serum lycopene, beta-carotene and ascorbic acid levels.

Intervention

Participants in the experimental group were assigned to receive tomato with mixed fruit juice (DOI KHAM 98% Tomato Juice with mixed fruit juice mocktail—200 ml). One box of the juice contained tomato juice 64%, strawberry juice 20%, passion fruit juice 20% and orange juice 4%. The ingredients per box were energy 100 kcal, protein 1 gm, total carbohydrate 23 gm (8% of Thai RDI), fiber 5 gm (20% of Thai RDI), sugar 11 gm, maltitol 1.04%, sodium 55 mg (2% of Thai RDI), Vitamin A (25% of Thai RDI), lycopene 60 mg, vitamin C (30% of Thai RDI), calcium ( 2% of Thai RDI), and iron ( 2% of Thai RDI). We recommended a daily dose of 100 ml per day after the morning meal in order to increase absorption of lycopene. This product was approved by the Thai and international FDAs [45]. Both experimental and control groups were prescribed with usual care as according to the regular visit by their physician. Possible undesirable effects from the mixed juice were noted and reported to an interviewer at the 2nd visit.

Data and statistical analysis

Data were processed with IBM SPSS statistics 23 software. The statistical analysis was performed by using program R version 4.1.1. An intention-to-treat analysis was conducted for this study. Participants’ socio-demographic data and clinical characteristics were analyzed using descriptive analysis. Continuous variables were presented as means and standard deviations (SD) for normal distributions. Ordinal and nominal variables were presented as numbers and percentages. For statistical analysis, we used the Student’s t-test for continuous variables, \(\chi\)2 test for categorical variables, and Fisher's exact test for nonparametric categorical data.

The correlation between frequency of antioxidant-rich diet consumption and level of asthma control was analyzed by using the Fisher's exact test (nonparametric categorical data). In addition, the association between serum antioxidant levels and level of asthmatic control was analyzed using the Student’s t-test. The mean difference between the 1st and 2nd visits in each group for pulmonary function tests, ACT, PAQLQ, ICS dose, serum antioxidants were calculated using the paired t-test (parametric statistics). The Student's t-test was used for comparing mean changes between control group (D1) and experimental group (D2). P-values of less than 0.05 were defined as statistically significant.