Efficacy of Ginger as Antiemetic in Children With Acute Gastroenteritis

A Randomised Controlled Trial

Rita Nocerino; Gaetano Cecere; Maria Micillo; Giulio De Marco; Pasqualina Ferri; Mariateresa Russo; Giorgio Bedogni; Roberto Berni Canani

Disclosures

Aliment Pharmacol Ther. 2021;54(1):24-31. 

In This Article

Methods

Trial Design

This randomised, double-blind, parallel-arm, placebo-controlled trial was performed in collaboration with family pediatricians of the Naples city area working for the Italian National Health System and was coordinated by the Department of Translational Medical Science of the University Federico II. The trial was approved by the Ethics Committee of University Federico II of Naples and was performed in accordance with the Helsinki Declaration (Tokyo revision, 2004), and with the pertinent European and Italian regulations about privacy. Written informed consent to participate in the study was obtained by the parents of the children. The trial was registered on https://clinicaltrials.gov/ with the identifier NCT02701491. Such registration was performed before the enrollment of the first patient.

Participants

The inclusion criteria were age between 1 and 10 years; suspected AGE-related symptoms lasting <12 h: AGE-associated vomiting (not bilious or bloody) from <4 h; modification of stool pattern lasting <12 h; mild to moderate dehydration evaluated as described elsewhere.[29]

The exclusion criteria were concomitant presence of other diseases, including neurologic and neuropsychiatric diseases; genetic and metabolic diseases, autoimmune diseases, immunodeficiencies, celiac disease, cancer, adverse food reactions (including ginger allergy); functional gastrointestinal disorders; inflammatory bowel diseases; liver diseases; pancreatic diseases; malformations of the gastrointestinal tract; infectious diseases other than AGE; severe dehydration; malnutrition defined as weight-for-height <3 standard deviation scores (SDS); previous surgery of the respiratory, gastrointestinal or urinary tract; use of gastric acidity inhibitors, antibiotics, antiemetics or other drugs in the 2 weeks before the enrollment; use of prebiotics, probiotics or symbiotics in the 2 weeks before the enrollment; participation to other studies.

We evaluated socioeconomic status according to the Hollingshead Four Factor Index for Socioeconomic Status, which is a validated tool to assess this variable.[30]

The final diagnosis of AGE was reached in the presence of ≥3 bowel movements of soft or liquid stools over 24 h, with or without fever, following the guidelines of the European Society for Pediatric Gastroenterology Hepatology and Nutrition and of the European Society for Pediatric Infectious Diseases.[31] Microbiological and other laboratory investigations were performed only for specific clinical reasons.

Intervention

The placebo and ginger products were in liquid form and their composition is given in Table 1. Products were produced under Good Manufacturing Practice by the Laboratory of Budetta Farma (Montecorvino Pugliano, Italy). Only a single batch of both ginger and placebo were used in the whole study. The distribution of the treatments was carried out by the coordinator center (Department of Translational Medical Science at the University Federico II, Naples, Italy). The packaging, color, weight, smell and taste of the ginger and placebo were similar. The similar smell and taste were obtained by adding anise and aromas to both products. The first dose of treatment (20 drops containing 10 mg of product) was administered by the family pediatricians immediately after the enrollment of the children and was followed by the administration of hypotonic oral rehydration solution (ORS) after 30 min.[31] The parents were instructed to administer 20 additional drops of treatment every 8 h after the first dose, until the resolution of vomiting. Additional doses were provided only in the presence of ≥1 episode of vomiting in the previous 8 h. The parents received a glass bottle with a pipette and a recyclable cardboard case and were instructed by their family pediatricians on how to use it. In addition, the parents were given a daily diary and were instructed by the family pediatricians on how to compile it. The diary recorded: quantity of ORS; whether ORS was refused; number of episodes of vomiting; presence of diarrhea or abdominal pain; number of bowel movements; stool consistency (Bristol stool scale); presence of systemic symptoms such as fever, headache, and irritability; suspected adverse reactions; hospitalisations; use of intravenous fluid therapy; number of days of school lost by the children; number of days of work lost by the parents. A complete medical examination was performed every 24 h by the family pediatricians until the disappearance of all AGE-related symptoms. Unscheduled visits were performed if necessary.

Outcomes

The primary outcome was the occurrence of ≥1 episode of vomiting after the first dose of ginger or placebo administered by the family pediatricians. The secondary outcomes were the incidence and the number of episodes of vomiting 24 h (day 1) and 48 h (day 2) after the first dose of treatment. Other outcomes were the quantity of ORS taken by the children in the 4 h after the first dose of treatment; the number of children refusing ORS; the number of children with diarrhea at 24 h (day 1), 48 h (day 2) and 72 h (day 3) from the first dose of treatment; the number of children requiring intravenous fluid rehydratation; the number of children requiring hospitalisation; the number of children not attending school at 24 h (day 1), 48 h (day 2) and 72 h (day 3) from the first dose of treatment.

Sample Size Calculation

Seventy-three subjects per arm were needed to detect an absolute difference of 20% (from 35% to 15%) in the occurrence of ≥1 episode of vomiting after the first dose of treatment between the placebo arm and the ginger at an alpha level of 0.05 with a power of 0.80 (Pearson's chi-squared test) (Stata 14.0, Stata Corp, College Station, TX, US). Such difference was considered clinically relevant in another study.[29] Estimating a dropout rate <3% basing on the results of a previous study,[29] we enrolled 75 subjects per arm, for a total of 150 subjects.

Randomisation

The family pediatricians administered the treatment according to a computer-generated stratified randomisation list produced using the ralloc command[32] (Stata 14.0, Stata Corp, College Station, TX, USA). The randomisation list employed 5 strata, 1 for each family pediatrician involved into the study, and block sizes of 2. Thus, each pediatrician had her/his own randomisation list involving 30 children randomly assigned to ginger or placebo in 1:1 ratio using block sizes of 2.

Allocation Concealment

Ginger and placebo were packaged in glass bottles and consecutively numbered according to the randomisation list generated for each family pediatrician (see Section 2.6.).

Blinding

The family pediatricians (assessors of the primary outcome), the children and their parents (assessors of the secondary and other outcomes), and the researchers who performed data entry, were blinded to the treatment. The statistician who performed the analysis was not blinded to the treatment.

Data Collection

Study monitoring was performed by an independent clinical trial monitor and included on-site visits and telephone interviews with family pediatricians. The clinical trial monitor reviewed the clinical forms for completeness, clarity, and consistency. All the data were recorded anonymously and entered into the study database by the same researcher. The study database underwent data cleaning according to standard procedures and was locked before statistical analysis.

Compliance

To assess compliance to the treatment, the parents were asked to return the bottles containing the treatment. Compliance to the treatment was defined as the consumption of 100% of it.

Statistical Analysis

Most continuous variables had non-Gaussian distributions, and all are reported as medians and interquartile ranges (IQR). Discrete variables are reported as numbers and proportions. The SDS of weight, height, and body mass index (BMI) were calculated using the World Health Organisation reference data.[33]

The primary outcome, that is, the occurrence of ≥1 episode of vomiting after the first dose of treatment, was evaluated using a binomial regression model. The response variable of the model was the occurrence of ≥1 episode of vomiting after the first dose of treatment (discrete: 0 = no; 1 = yes) and the predictor variable was treatment (discrete: 0 = placebo; 1 = ginger). The point estimate and the 95% confidence interval (95% CI) of the absolute risk reduction (ARR) were obtained from the model.[34] The 95% CI of the number needed to treat (NNT), that is, the number of patients to treat to prevent the occurrence of ≥1 episode of vomiting after the first dose of treatment, was calculated using Bender's formula.[35] We performed an intention-to-treat analysis (ITT) of the primary outcome by considering the children lost after randomisation as follows: (1) all missing values of the primary outcome set to the worst outcome in both the ginger and placebo arms (equal-case scenario ITT) and (2) missing values of the primary outcome set to the worst outcome in the ginger arm and to the best outcome in the placebo arm (worst-case scenario ITT).[36] The worst outcome was defined as the occurrence of ≥1 episode of vomiting after the first dose of treatment; the best outcome was its opposite. The equal-case scenario ITT for the primary outcome was prespecified by the study protocol, as per standard practice. The worst-case scenario ITT for the primary outcome was implemented post hoc.

The secondary outcomes, which involve repeated measures, were evaluated using per-protocol analysis. The incidence rate of vomiting after 24 h (day 1) and 48 h (day 2) from the first dose of treatment was evaluated using a binomial regression model for repeated measures.[34,37] The response variable of the model was the incidence of vomiting (discrete: 0 = no; 1 = yes), and the predictors were treatment (discrete: 0 = placebo; 1 = ginger), time (discrete: 0 = day 1; 1 = day 2), and a treatment × time (discrete × discrete) interaction. Repeated measures were taken into account by using subject-specific cluster confidence intervals. Two prespecified between-group (ginger vs. placebo) within-day (day 1 and day 2) contrasts were used to calculate the time-specific ARR. ARR and p values were corrected using a Bonferroni correction for two contrasts (day 1 and day 2).

The number of episodes of vomiting were calculated using a negative binomial regression model for repeated measures.[34,37] The response variable of the model was the number of episodes of vomiting (count) and the predictors were treatment (discrete: 0 = placebo; 1 = ginger), time (discrete: 0 = day 1; 1 = day 2), and a treatment × time (discrete × discrete) interaction. Repeated measures were taken into account by using subject-specific cluster confidence intervals. Two prespecified between-group (ginger vs. placebo) within-day (day1 and day 2) contrasts were used to calculate the time-specific ARR. ARR and P values were corrected using a Bonferroni correction for two contrasts (day 1 and day 2).

The remaining outcomes were reported only as descriptive (and not inferential) statistics, using medians and interquartile ranges (IQR) for continuous variables and numbers and proportions for discrete variables. Such outcomes were: quantity of ORS assumed by the children in the 4 h after the first dose of treatment; number of children refusing ORS; number of children with diarrhea after 24 h, 48 h and 72 h from the first dose of treatment; number of children requiring intravenous fluid therapy; number of children requiring hospitalization; number of children not attending school after 24 h, 48 h and 72 h from the first dose of treatment.

Statistical analysis was performed using Stata 16.1 (Stata Corporation, College Station, TX, USA).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....