Patients and ethical issues
The present study enrolled outpatients who underwent sedative endoscopy at the National Hospital Organization Ureshino Medical Center from two prospective studies that analyzed the usefulness of MPADSS and the modified Aldrete score, respectively, as discharge standards. Outpatients > 20 years of age who underwent esophagogastroduodenoscopy, colonoscopy, or endoscopic ultrasonography under sedation were candidates for the present study. Intravenous sedation was not generally used in patients allergic to midazolam or in pregnant or breastfeeding women, thus excluding these patients. Outpatients who underwent therapeutic endoscopy were also excluded from the present study.
This study was conducted in accordance with the tenets of the Declaration of Helsinki and the guidelines of the Consolidated Standards of Reporting Trials (CONSORT). The study protocol and the consent procedure were approved by the Ethics Review Committee of the National Hospital Organization Ureshino Medical Center (approval number 19–03), and the study was registered with the University Hospital Medical Information Network (UMIN) Clinical Trials Registry (UMIN000037259) on 3 July 2019.
Sedation and monitoring
Sedative and analgesic drugs were generally used as follows: An initial bolus of midazolam (3 mg for patients with a bodyweight of < 50 kg and 4 mg for patients with a bodyweight of ≥ 50 kg) was administered through an intravenous catheter. When the patients showed signs of discomfort, restlessness, agitation, and/or a response to verbal commands, 1 mg of midazolam was added as appropriate [5]. Analgesic agents (7.5 mg of pentazocine or 17.5 mg of pethidine hydrochloride) were also used if necessary. Analgesic drugs were selected by each endoscopist, considering the patient’s age or physical condition.
Vital signs, namely blood pressure, heart rate, and blood oxygen saturation were recorded before the induction of sedation. During the endoscopy, vital signs were monitored every 5 min. When oxygen saturation was < 92%, nasal oxygen supplementation (2 L/min) was administered. When a patient’s vital signs fluctuated by ≥ 20% compared with the baseline values, the endoscopic procedure was temporarily stopped until the values returned to baseline values.
Flumazenil, a benzodiazepine antagonist, was administered after the endoscopy as necessary.
Assessment of the patients’ conditions after sedative endoscopy
Each patient’s clinical condition after sedative endoscopy was assessed by nurse-administered MPADSS as the discharge standard in 181 outpatients from July 2019 to January 2020 (group M). The patients’ conditions were assessed using a nurse-administered modified Aldrete score in 195 outpatients from July 2020 to December 2020 (group A). In both studies, the patients’ conditions were determined 60 min after the endoscopy. When the MPADSS or the modified Aldrete score reached ≥ 9 (maximum score: 10), patients were allowed to discharge from the hospital. Otherwise, the assessment was repeated every 30 min. When the patients did not reach a MPADSS or modified Aldrete score ≥ 9 3 h after endoscopy, the patient was admitted to the hospital.
Data collection
The following patient demographic data were collected by reviewing the patients’ electronic clinical records: age, sex, alcohol consumption, smoking habit, body mass index, American Society of Anesthesiologists physical status (ASA-PS) classification, previous history of endoscopy, and comorbidities (including the Charlson comorbidity index score). Endoscopy-related data, namely the types of endoscopy (upper or lower gastrointestinal endoscopy, or endoscopic ultrasonography) and endoscopic outcomes (procedure time, types and amount of sedative, analgesic dose, antagonist, and endoscopist specialty), and peri-procedural adverse events (unstable vital signs indicated by ≥ 20% decrease from the baseline values) were collected by chart review.
After sedative endoscopy, the MPADSS or the modified Aldrete score at 60 min, recovery time, and vital signs before discharge were measured. Any adverse events before discharge and the need for hospitalization were recorded. Adverse events within 24 h after discharge were investigated by telephone questionnaire the day after endoscopy.
Study endpoints
The primary endpoint of the present study was the difference in the discharge time after sedative endoscopy between the two groups. The secondary endpoint was the difference in the adverse events rate before discharge, need for hospitalization, and adverse events within 24 h after discharge between the two groups.
Sample size calculation and statistical analysis
The sample size was estimated based on previously published results [17] and on the results from the previous data conducted from endoscopy in our hospital from July 2018 to June 2019. Based on these studies, we assumed that the rate of recovery time within 60 min after endoscopy to be 45%. We hypothesized that a difference of 20% in the rate of recovery time between the two groups would constitute a clinically meaningful difference. Assuming a power of 80% and an alpha of 0.025 (one-sided), at least 214 patients (107 patients in each group) would be required in this study. Assuming that 10% of patients could be lost to follow-up, a sample size of 236 patients (118 patients in each group) was planned.
Propensity score matching analysis was used to assess the clinical usefulness of MPADSS and the modified Aldrete score as the discharge criteria. This method was used to adjust for significant differences in the patients’ baseline characteristics and to minimize the influence of possible confounding factors [18, 26]. The two groups were matched at a 1:1 ratio (120 patients in each group) with adjustment for seven covariates (age, sex, ASA-PS classification, comorbid malignant diseases, procedure time, mean midazolam dose, use of analgesics) to minimize inherent bias (Fig. 1). These seven covariates were selected based on the opinions of expert endoscopists (DY, TM, NH, YT, and ST). This model yielded a C statistic of 0.754, indicating a preferable ability for the comparison between groups M and A. The caliper width of the propensity score matching was 0.20.
Categorical data were expressed as frequency and percentage, and the chi-square test was used to identify differences between the two groups. Normally-distributed numerical data were expressed as mean ± standard deviation, and Student’s t-test was used to determine differences between the two groups. Numerical data with a skewed distribution were expressed as median [interquartile range], and the Mann–Whitney U-test was used for comparisons between the groups. Levels of significance for all comparisons were reported, regardless of statistical significance, as P values or confidence intervals. A P value < 0.05 was considered statistically significant for each test. All statistical analyses were performed with JMP, version 13.2.0 (SAS Institute Inc., Cary, NC, USA).