【Abstract】 Objective The present study was designed to compare safety and efficiency of deep sedation and moderate sedation during upper gastrointestinal endoscopy in inpatients having comorbidities and classified ASA class Ⅱ-Ⅲ. Methods A prospective, randomized and double-blinded study was performed. One-hundred and twenty inpatients were enrolled, with 40 patients in each group. Inpatients with comorbidities and ASA class Ⅱ-Ⅲ were randomized into moderate sedation (Group M, midazolam+sufentanil), deep sedation with two anesthetics (Group D2, propofol+sufentanil) and deep sedation with three anesthetics (Group D3, propofol+midazolam+sufentanil). The vital signs, sedation induction time, procedure time, recovery time, patients’ comfortability and satisfaction, endoscopists’ satisfaction, sedation-related complications and procedure-related complications were recorded. Results There was no statistical difference in the three groups regarding age, gender, body mass index, ASA classification, or procedure time. Moderate sedation could significantly shorten induction time (P=0.001) and recovery time (P=0.002). Compared with group M, patients in group D2 and group D3 had higher risks of sedation-related adverse events. Endoscopists were satisfied with the different levels of sedation in all three groups. Patients in group D2 and group D3 had more comfortability (P<0.001). Compared with group D2, patients in group D3 required significantly less propofol to reach deep sedation (P<0.001), and it resulted in a lower incidence of circulation and respiration adverse events. Conclusion For most inpatients with comorbidities and classified ASA class Ⅱ-Ⅲ, routine gastrointestinal endoscopy can be performed safety and efficiency with either moderate or deep sedation. For the patient with unstable cardiopulmonary status, moderate sedation might be a better choice instead of deep sedation.

【Key words】 Upper gastrointestinal endoscopy; Moderate sedation; Deep sedation; Comorbidities

1 Background

Nowadays, it is a routine practice in patients undergoing endoscopy to administer intravenously sedative medication. Sedation is to relieve patient anxiety and discomfort, improve the outcome of the examination, and diminish the patient’s memory of the event[1].Although endoscopic procedures can be performed without sedation, the use of sedation is associated with a higher satisfaction of patients[1-2]. Demand for gastrointestinal endoscopic procedure sedation is expanding. A research found that anesthesia service for gastroenterology procedures accounted for 33.7% in 2010 and 47.6% in 2013 in Medicare patients, and 38.3% in 2010 and 53.0% in 2013 in commercially insured patients[3].

Sedation comprises a continuum of states that include minimal sedation (anxiolysis), moderate (conscious) sedation, deep sedation, and general anesthesia (Table 1)[1]. Under moderate sedation, patients continue to respond purposefully to verbal commands, either alone or with light tactile stimulation, and no interventions are needed to maintain a patient airway or spontaneous ventilation. Under deep sedation, patients can not be aroused easily but respond purposefully to repeated or painful stimulation. Deeply sedated patients may have inadequate spontaneous ventilation and may require assistance to maintain a patent airway[1, 4-6].

Table 1 Levels of sedation and anesthesia

Modified from Guidelines for sedation and anesthesia in GI endoscopy.

Routine endoscopy can be performed successfully with either moderate or deep sedation[6-8].Traditional moderate sedation utilizes benzodiazepines alone or in combination with an opiate analgesic. Its safety profile is reflected in its general acceptance as the standard regimen for sedation in general gastrointestinal endoscopy. Furthermore, this regimen is often used as the standard for comparison against novel sedative combinations or compounds being evaluated for use during endoscopy[9]. Deep sedation provides patients with much more comfortability, but it results in a significant incidence of a series of complications, such as hypotension, hypoxemia, apnea and other adverse events[10-11]. Nonetheless, some studies suggested that no significant difference in terms of complication incidence between deep sedation and moderate sedation[7, 12].To date, the literature is still expanding with special regard to the deep sedation safety during the gastrointestinal endoscope.

At present, numerous studies proved that both deep sedation and moderate sedation were available to achieve a safe, comfortable and technically successful endoscopic procedure.However, there is no study to investigate the sedation regimen on patients with comorbidities and classified as American Society of Anesthesiologist (ASA) class Ⅱ or Ⅲ. For the patients with comorbidities like cardiopulmonary, liver, renal disease or neurological disorders, pharmacodynamic and pharmacokinetic properties of the sedatives and analgesics may be not similar as that in the healthy patients. Meanwhile, those patients with comorbidities may have different tolerance to different levels of sedation. The purpose of this study is to compare the safety and efficiency of moderate sedation and deep sedation in ASA class Ⅱor Ⅲ inpatients with comorbidities.

2 Methods

2.1 Ethics Ethical approval for this study was provided by the Biological-Medical Ethical Committee of West China Hospital, Sichuan University. Informed consent was obtained from each patient or their designated representatives at preoperative visit.

2.2 Protocols Before the trail, enrolled patients were randomized into 3 groups: moderate sedation (Group M, midazolam + sufentanil), deep sedation with two anesthetics (Group D2, propofol+sufentanil) and deep sedation with three anesthetics (Group D3, propofol+midazolam+sufentanil). Randomized allocations were generated using a computer random number generator. To standardize technical skills and experience, every case was performed by the same anesthesiologist and board certificated endoscopists. None of the anesthesiologist participated in the data assessment or analysis. Patients, investigators, endoscopists and individuals participating in data analysis were all blind to group allocation.

2.3 Patients A prospective, double-blinded, and randomized controlled trail was conducted on inpatients undergoing analgesia gastroscopy. Inclusion criteria were age greater than 18 years, ASA physical status Ⅱ or Ⅲ inpatients with comorbidities such as hypertension, diabetes mellitus, coronary heart diseases (CHD), chronic obstructive pulmonary disease (COPD), anemia, liver cirrhosis, serous cavity effusion. Exclusion criteria included patients older than 80 years, pregnancy, a history of allergy to propofol and its components, serious COPD with respiratory failure, chronic cardiac dysfunction with New York Heart Function Assessment (NYHA) classification level 4 or above or acute heart failure, serious anemia, serious serous cavity effusion which have effect on the respiratory or circulatory function, undergoing therapeutic esophagogastroduodenoscropy (EGD) at the same time and the patients who can’t tolerate the anesthesia or the endoscopic procedure.

2.4 Anesthesia and Surgical Procedure All enrolled patients fasted for 8 hr before the procedure and received no premedication. After topical anesthesia of the oropharynx with Lidocaine Hydrochloride Gel, the patients lied on their left side on the operation bed. During sedation administration each patient’s vital signs were recorded by an investigator, including Non-invasive arterial blood pressure, electrocardiography (Lead Ⅱ), and pulse oximeter oxygen saturation (SpO2). Oxygen (4 L/min) was administrated via facemask.

Patients in group M received a bolus of 0.02-0.03 mg/kg midazolam and 0.1μg/kg sufentanil to reach the moderate sedation. Patients in group D2 were administrated with 0.1μg/kg sufentanil and propofol (10 mg/mL) with a bolus of 0.5 mg/kg at a rate of 0.5 mL/s, and then propofol was given at a rate of 0.25 mL/s until the patients demonstrated the signs of deep sedation. For the patients randomized to D3 group, both midazolam and sufentanil were given at the initiation of sedation in the same manner as in the M group. Thereafter, an initial bolus of propofol was given in the same manner as in the D2 group to reach the same end-point. In an effort to avoid the pain on injection, lidocaine 1mL was added to the propofol immediately before administration.

Sedation level was measured according to practice guidelines for sedation and analgesia by nonanesthesiologist as shown in table 1. “Response”was defined as any verbal or purposeful physical reaction to a verbal or tactile stimulus. The observer spoke the patient’s name first with normal voice, then with a louder voice. If the patient failed to respond, a light tactile stimulus on the shoulder was provided, followed within several seconds, if necessary, by a second light stimulus. Patients who failed to respond to these 4 stimuli were considered to be deeply sedated. No communication was permitted between the endoscopy staff and the observer during the examination, and all results were concealed from the other investigators until completion of patient enrollment.

All the patients were performed by the experienced endoscopists. After the procedure, patients were transferred to the recovery area and evaluated every 5 min by the Post Anesthesia Discharge Scoring System(PADSS)(Table 2)[13]by an observer not included in the study and blinded to the patients’ randomization. The achievement of ten-point value on the PADSS was considered adequate recovery for discharge from the Endoscopy Unit.

Before discharge, the patient was required to rate his satisfaction with the procedure on a 10-cm visual analog scale (VAS, 0=minimum, 10=maximum satisfaction), together with his comfortability. Furthermore, the endoscopist was asked to evaluate the satisfaction with the different levels of sedation for completion of the procedure.

2.5 Outcome

2.5.1 The vital signs including systolic blood pressure (SBP)/diastolic blood pressure (DBP)、HR、SpO2、

Table 2 Postanesthesia recovery score

Modified fromNayar DS1, Guthrie WG, Goodman A, et al. Comparison of Propofol Deep Sedation Versus Moderate Sedation During Endosonography

RR was recorded at the following time points: baseline (T0), the time reaching the required sedation level (T1), 1 min after insertion of endoscope (T2), removal of endoscope (T3), arrival at the recovery area (T4), emergence from sedation (T5, the PADSS reached to 10 points).

2.5.2 Induction time (the time span between the start of an anesthetic infusion and the start of insertion of endoscopy), duration of procedure (from insertion of endoscope to removal of endoscope), recovery time (the time span between removal of endoscope and emergence from sedation), length of stay at the recovery area. The doses of each kind of anesthetics used in every patient were recorded.

2.5.3 The incidence of the procedure-related complications: nausea, vomit, cough, perforation of gastrointestinal tract, bleeding, and the times offailure in inserting the endoscope and the frequency of interruption of the procedure because of adverse events.

2.5.4 The incidence of sedation-related complications: hypotention (SBP<90 mmHg & MAP(mean arterial pressure)<60 mmHg), bradycardia (HR <50 bpm), desaturation (SpO2<90%), apnea (RR<8 /min & respiratory tract obstruction), faintness.

2.5.5 The application of rescue strategies: 0.5mg atropine was administrated when HR was less than 50 bpm. If the SBP was less than 80 mmHg or MAP less than 60 mmHg, patients were given metaraminol 1 to 2 mL at the concentration of 0.1 mg/mL. Hypoxemia was resolved with physical stimulation, airway manipulation (usually jaw thrust or nasopharyngeal airways), and delivery of high-flow oxygen by facemask.

2.5.6 Patient satisfaction and comfortability was measured by VAS, and endoscopists’ satisfaction was evaluated.

2.6 Statistical Analysis Statistical analysis was performed using the Statistical Package for the Social Sciences software (SPSS version 22 for Windows; SPSS Korea, Seoul, Korea). ANOVA or repetitive measurement deviation analysis were used for comparison of continuous data followed by Tukey’s multiple comparison. Continuous data was presented as mean ± SD. Categorical data were presented as numbers (percentage) and analyzed by chi-square test. Statistical significance was considered whenP<0.05.

3 Results

3.1 Study population and baseline characteristics A total of 120 consecutive inpatients scheduled for upper gastrointestinal endoscopy were studied, with 40 in each group. The baseline characteristics of patients were no difference among the 3 groups (Table 3), including age, gender distribution, body weight, ASA classification and comorbidities.

3.2 Propofol used and different duration tested in the procedure A summary of the mean total dose of propofol administered and the mean procedure-related duration is presented in Table 4. The mean dose of propofol used to reach deep sedation was 71.5 mg and 41.25 mg in D2 group and D3 group respectively, with propofol in D3 group much less than in D2 group. The mean endoscopic procedure duration was not statistically different among the 3 groups. Compared with M group, the recovery time and length of stay at recovery area in D2 and D3 group were much longer (P<0.05). However, as for the induction time, D2 group was much longer than M group and D3 group, with no difference between M group and D3 group.

3.3 The changes of MAP, HR, SpO2during the endoscopic procedure The three groups were similar with respect to the baseline MAP and HR. The patients in the three groups had the similar MAP when they left the recovery room (T5) as their baseline MAP (T0) (P>0.05). From reaching the required sedation level to emerging from sedation (from T1 to T4), the patients in group D2 and D3 demonstrated lower MAP than in group M. No difference was found between group D2 and group D3. Although MAP in three groups was all more than 60 mmHg, MAP in group D2 and D3 fluctuated more than in group M.

Table 3 Baseline characteristics of study

Table 4 Propofol used and different duration tested in

HR in each group at different time point was more than 60. Compared with group M, HR at T1, T2, T3 in group D2 were slightly slower. Except that, there was no difference among groups at other time points.

The baseline SpO2in group D2 and D3 was higher than in M group statistically, but had no clinical significance. However, SpO2in each group at different time point showed no difference (Table 5).

Table 5 The different changes in MAP, HR, SpO2 at different time

3.4 The overall rate of adverse events and support methods There was no severe complication associated with the endoscopic procedure in all three groups(Table 6). However, hypotension was the most frequently reported adverse event in deep sedation groups (group D2 and group D3) compared to moderate sedation group (group M). Only 4 patients in group M showed MAP decreased more than 20%, with 21 and 13 patients in group D2 and group D3 respectively, which showed significant statistical difference among groups. Meanwhile, group D2 had more patients with SBP < 90 mmHg than group M. Although the same trend occurred between group D3 and group M, it didn’t reach statistical difference. Correspondingly, the patients requiring vasoactive agents therapy in group D2 was more than group M. There was no difference in respiratory complications among 3 groups. However, no patients in M group experienced apnea, obstructive airway and SpO2<90%. Apnea was observed in 2 patients in D2 group and in 3 patients in D3 group respectively. Four patients in group D2 and 6 cases in group D3 had obstructive airway, with 1 patient SpO2<90% in each deep sedation group.

Table 6 The overall rate of adverse events and support methods

3.5 Postprocedure assessment Patients in D2 group and D3 group expressed more comfortability than in M group. However, both patients and endoscopists felt satisfied with different levels of sedation.

4 Discussion

Local anesthesia, moderate sedation and deep sedation have been used in gastrointestinal endoscopic procedure. Although there are abundant studies comparing moderate sedation versus deep sedation in patients undergoing endoscopy, existing literature does not adequately study the safety of sedation on patients having comorbidities. The results of our study provide the preliminary evaluation on the safety of sedation on those patients.

In this study, we used one moderate sedation regimen (midazolam+sufentanil) and two other deep sedation regimens (propofol+sufentanil and propofol+midazolam+sufentanil). We found the sedation-related cardiopulmonary complications showed significant difference among groups. More patients in deep sedation groups reported hypotension (21 in group D2 and 13 in group D3) than moderate sedation group. Meanwhile, the MAP in group D2 and D3 fluctuated more than that group M. The hypotension complication might be ascribed to the effects of vasodilation and myocardial depression by propofol. In spite of MAP in all groups more than 60 mmHg, this gave us enough hints that moderate sedation might be a safer and better choice for patients with relatively hypotension at the baseline.

There was no respiration-related complication in Mgroup. Nonetheless, some patients in deep sedation groups demonstrated apnea and airway obstruction. One patient in group D2 and one in D3 showed transient hypoxemia. Similar results were obtained by some other studies[14]. Despite the known potential of propofol for inducing cardiorespiratory suppression, the pooled incidence of hypoxemia (11%) and hypotension (5%) was similar to pooled results from sedation with midazolam alone and combined midazolam and narcotics. This corroborated safety information from 2 large uncontrolled observational studies of propofol sedation in which the incidence of hypoxemia (<85% hemoglobin saturation) was only 0.25% and the incidence of significant respiratory compromise (prolonged apnea, laryngospasm, aspiration) was<0.1%[15-16].

The recovery time and length of stay at recovery area in moderate sedation group was less than in deep sedation groups, and the induction time in D2 group was much longer than in M group and D3 group, with no difference between M group and D3 group. This result in our study is not consistent with some other research results[13, 17]. As some research reported, propofol sedation is used increasingly in clinical practice in part because providers believe that it can shorten sedation and recovery time and thereby enhance efficiency of the endoscopy unit[18-19]. Pooled data suggest that recovery time was shorter for both EGD and colonoscopy in patients sedated with propofol than with midazolam with or without narcotics and time to sedation for colonoscopy also was shorter with propofol than with midazolam with or without narcotics. Several possible factors might be taken into account to this inconsistency: ①compared with M group, propofol was additionally used except the same dose of midazolam and sufentanil as in M group. ② in order to provide a smooth induction process for the patients with comorbidities and classified ASA class Ⅱ or Ⅲ, the propofol was administrated at a constant rate, which contributed to the longer induction time in D2 and D3 group. ③in our gastrointestinal endoscopy unit, after fully recovery, the patients, without the need to walk to leave, were escorted to the ward by transport workers. As a result, the patient stayed at the recovery area just for a short time.

Although patients in deep sedation group felt more comfortable than patients in moderate sedation group, all the patients expressed the equal level of satisfaction with the operation. These data suggest that procedure comfortability alone might not negatively affect the patients’ satisfaction with the procedure. Nevertheless, patients undergoing therapeutic EGDs or colonoscopies such as polyp resection and endoscopic mucosal stripping require deep sedation to maximize patient cooperation and the likelihood of a successful examination. For the patients with comorbidities, using multi-anesthetic like midazolam and narcotic can reduce the dose of propofol so that the dose-related propofol complications including hypotension, respiratory depression and bradycardia are less likely to occur[16, 20]. It has also been demonstrated in our study that the dose of propofol used in D3 group was much less than in D2 group. In other words, propofol combined with midazolam and narcotic may be the better choice for patients with comorbidities undergoing therapeutic EGDs and coloscopies.

There were two limitations in our study: ①we did not compared medication dosage between cases associated with adverse events and cases without comorbidity.②both the severity of comorbidity and its relationship with the sedation-related complication were not defined so that we could not distinguish whether the adverse events were caused by the anesthetics used or by patients’ comorbidities.

5 Conclusion

For most inpatients with comorbidities and classified ASA class Ⅱ-Ⅲ, routine upper gastrointestinal endoscopy could be performed safety and efficiency with either moderate or deep sedation. Forpatients with unstable cardiopulmonary status, moderate sedation might be a better choice instead of deep sedation. It is possible that changing the concentration and the ratio of sedatives and analgesics may lead to an optimal sedation regime for patients with comorbidities, which need further investigation.