Efficacy of Carbonated Beverages in the Resolution of Alimentary Esophageal Impaction: a Multicenter Randomized Clinical Trial
Efficacy of Carbonated Beverages in the Resolution of Alimentary Esophageal Impaction: a Multicenter Randomized Clinical Trial
Objectives: To evaluate the efficacy of the administration of carbonated beverages in the resolution of dietary esophageal impaction, and to identify subgroups in which their effectiveness may be different: patients with/without esophageal diseases, with/without a history of impaction.
Methods: A clinical trial was conducted randomizing consecutive patients with a clinical diagnosis of dietary esophageal impaction to receive 200cc of carbonated drink or not at all, excluding patients with a previous diagnosis of non-esophageal stenosis that could not be overcome with a gastroscope and suspected impaction by non-alimentary foreign bodies, spines or bones. Subsequently, an endoscopic evaluation will be carried out in all cases, checking the presence or absence of a foreign body and removing it if necessary, as well as assessing and recording the possible underlying esophageal pathology.
A telephone follow-up will be carried out after 7 days, recording possible adverse events.
METHODOLOGY
Design: Randomized multicenter clinical trial.
Period: patients treated in the Emergency Department with a clinical diagnosis of esophageal foreign body impaction (cervical or retrosternal foreign body sensation, hypersalivation, inability to swallow) will be collected with indication by clinical practice by the emergency physician who evaluates them for upper gastrointestinal endoscopy.
Population selection:
Scope: patients treated in the Emergency Department of the participating hospitals.
Inclusion and exclusion criteria:
3.3. Recruitment procedure: all patients treated in the Emergency Department of the aforementioned hospitals with a clinical diagnosis of esophageal food impaction who meet the inclusion criteria and do not meet the exclusion criteria will be invited by the Emergency Department to participate in the study.
3.4. Sample size Most of the studies on carbonated beverages date from the 1980s and 1990s, are very short series, and in some of them they are combined with the administration of glucagon. In all of them, the success rate is 100%.
In the largest and most recent series of the studies reviewed, the effectiveness rate of the carbonated drink is 59%, which in the light of clinical experience may be exaggerated (possible bias in the selection of the most favorable patients). If no treatment is performed, the foreign body can be found in 80-90% of cases. Thus, considering a difference of 20% relevant, assuming a success rate in the intervention group of 35%, with a risk α=5% and a power of 80%, a total of 84 patients per group would be needed. Since the study has no follow-up, we consider it unlikely that there will be any losses.
In light of the results of our placebo-controlled clinical trial on the efficacy of glucagon, in which the only difference detected was a decrease in the complexity of endoscopy in the intervention group in patients with no personal history of impaction, it would be interesting to have a number of such patients in order to be able to detect differences with adequate power. According to our study, 47% of patients (66/140) have no history of impaction that required medical assistance. Thus, if we consider the n properly calculated as patients with no history, we need a total of 177 patients per group.
Intervention 4.1. Randomization It will be carried out once the patient agrees to participate in the work and the absence of exclusion criteria is verified. A stratified randomization will be carried out by center, so that each participating center will have its sealed randomization sequence. Within each stratum, pure randomization will be performed using a computer-generated sequence where the probability of belonging to each group will be 0.5. A total of 100 numbered opaque sealed envelopes will be available for each centre and will be stored in the emergency department and opened consecutively as patients are included.
4.2. Intervention The study will be completed after the formalization of the informed consent in accordance with Law 41/2002 on patient autonomy. Participation in the study will not alter your relationship with your doctor.
4.2.1 Pre-endoscopic evaluation
Patients will be identified in the emergency department. Those who meet the inclusion criteria and do not present any exclusion criteria will be invited to participate in the study. Once they consent to their participation, the endoscopist on duty will be contacted to plan the examination, which should be carried out in 30-120 minutes after the call in both groups. Once the endoscopic examination has been scheduled, the patient will be randomized. Based on this, 2 groups are established:
Data collection 5.1 Description of variables
5.2. Data sources Variable variables related to the patient's personal history will be collected in the form of an interview with the patient by the emergency physician, focusing on the presence of esophageal pathology, medications that affect esophageal motility and history of impactions, time of impaction and water intake after obstruction.
After performing the endoscopy, the physician who performs it will collect technical details such as the endoscopy time, the tools used for deimpaction, the type and location of the foreign body and the underlying esophageal pathology 5.3. Existence of a blind person It is not possible to conduct a double-blind study as the intervention (administration of carbonated beverage) cannot be compared with placebo. Patients in the Emergency Department will be instructed not to disclose to the endoscopist to which branch of the study they have been assigned.
Although the final report to be delivered to the patient at hospital discharge will include the treatment administered, since it is generated and delivered to the patient after the procedure has been performed. The evaluation of the efficacy of the treatment will be performed by the endoscopist during the endoscopic examination. Since the endoscopy unit does not have access to the software or treatment sheets of the emergency department, the endoscopists in charge of evaluating the outcome of the treatment will not be able to know the assigned group.
Follow-up 7-10 days after the endoscopic examination, a standardized telephone interview will be conducted to assess the possible presence of adverse effects. In case of the presence of symptoms, the causal relationship with the procedure will be evaluated by a group of 3 study investigators. The severity of the events will be assessed according to the usual classifications (CTCAE v.4) and according to the definitions of Royal Decree 1090/2015.
Analysis The statistical analysis of these will be carried out using the STATA program (StataCorp. Version 16. Stata Statistical Software. College Station, TX).
To verify that there are no imbalances between the experimental group and the control, means are compared with quantitative variables, assessing the magnitude of the association with Pearson's correlation coefficient (r). For binary categorical variables, proportions are compared, assessing the association with the OR obtained by logistic regression. For ordinal categoricals, the ROs are calculated with respect to the category determined as a reference.
For the descriptive variables, the arithmetic mean and the standard deviation will be calculated for the quantitative variables (variables that do not follow a normal distribution according to the Kolmogorov-Smirnov test will be described with median, minimum, maximum and interquartile range), and the qualitative variables will be expressed as percentages and their 95% confidence intervals.
The analysis of the primary endpoint, the proportion of patients without the presence of an impacted foreign body in both groups, will be compared using the homogeneity z-test without using the Yates correction. The confidence interval of the difference between the two groups will also be estimated. The analysis will be performed on an intention-to-treat basis and on a per-protocol basis (including only subjects who receive treatment within and perform the examination within the predetermined time frames).
Within the secondary endpoints, the proportion of complications in both groups will be checked by the χ2 test. The duration of the procedures will be compared using the Student's t-test. Multivariate logistic regression techniques will be used to evaluate the factors associated with the effectiveness of the treatment.
Novelty, originality This study continues a line of research established at our center with a previous clinical trial on the efficacy of glucagon with a sample size and a novel design, which provides data not demonstrated in randomized trials.
The efficacy of carbonated beverages for this indication has also not been proven with a study of the characteristics of the one we propose. Our findings would avoid unnecessary endoscopies, if the efficacy of these beverages is demonstrated, including them in the protocol of action in food esophageal impaction. Or, if, on the contrary, it were to prove ineffective, unnecessary waiting for endoscopy and inconvenience to the patient when attempting these futile measures would be avoided.
Inclusion Criteria:
Exclusion Criteria: