Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

Chang, 2015

Type of study: RCT

Setting and country: Taiwan

Funding and conflicts of interest: No financial support; no conflict of interest

Inclusion criteria: adult patients undergoing upper and lower gastrointestinal endoscopy

Exclusion criteria: an allergic reaction history to any of the study drugs, eggs, or soybeans; chronic exposure to opioid analgesics or sedative medication; inpatient status; a history of obstructive sleep apnea; seizure disorders; and gross obesity [a body mass index (BMI) greater than 42 in males or 35 in females).

N total at baseline:

Intervention: 50 patients

Control: 50 patients

Important prognostic factors²:

age ± SD:

I: 50.0 ± 13.5

C: 51.4 ± 14.2

Sex:

I: 46% M

C: 50% M

TCI for upper/lower gastrointestinal endoscopy

In the TCI group, the patients were given a dose of 2–2.5 mg of midazolam. Subsequently, we administrated propofol and alfentanil through an intravenous cannula utilizing a TCI system. The TCI system included two-unit infusion pumps controlled by a microprocessor system. The pharmacokinetic data distribution and elimination of propofol and alfentanil were programmed in the TCI system. We used the pharmacokinetic model proposed by Schnider et al. for propofol and the model proposed by Scott et al. for alfentanil. The patients received an initial target effect-site concentration

of 4 μg/mL for propofol until loss of consciousness

(loss of verbal contact) was achieved. This was immediately followed

by a target effect-site concentration of 2.5 μg/mL. During the endoscopy procedures, the maintenance concentration was adjusted to 0.5 μg/mL higher or lower when needed.

Manual administration of propofol for upper/lower gastrointestinal endoscopy.

In the control group, patients were administered a dose of 0.27 mg of alfentanil and a dose of 2–2.5 mg midazolam based on their age and body weight. After 3 min, propofol was administered and titrated to the desired sedation effect with a dose of 20–30 mg. If necessary, the nurse anesthetist empirically administered additional doses of alfentanil and propofol based on the patient’s response during the procedure.

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation

Not reported

Propofol consumption

I: 144.9 ± 27.1 mg

C: 118.3 ± 38.7 mg

(p<0.05)

Absolute mean difference: 26.60 mg [95%CI 13.50; 39.70]

Adverse events

Hypotension: Two patients in the TCI group were treated with low doses of ephedrine because of severe hypotension

Respiratory depression (Mean EtCO2 >60 mmHg or assisted ventilation (SpO2 <95%)):

I: 7 (14%)

C: 12 (24%)

(Overall p=0.44)

Patient satisfaction

Not reported

Author conclusions: In summary, our study showed that sedation using TCI for GI

endoscopy provided effective and safe sedation and was associated with better sedation quality. The depth of sedation appeared to be appropriate and allowed the patients to be easily maneuvered during the procedures. We believe that TCI can be used to provide routine sedation for patients undergoing GI endoscopy.

Chiang, 2013

Type of study: RCT

Setting and country: Taiwan

Funding and conflicts of interest: No funding reported, no conflict of interest

Inclusion criteria:

Patients undergoing bidirectional endoscopy and ASA physical status 1 or 2.

Exclusion criteria: a history of known allergy to propofol or its lipid emulsion, or drug or alcohol abuse

N total at baseline:

Intervention: 110

Control: 110

Important prognostic factors²:

age ± SD:

I: 50.51 ± 11.44 years

C: 51.14 ± 10.19 years

Sex:

I: 55% M

C: 54% M

TCI for bidirectional endoscopy

The propofol TCI group was set with the Model of TCI Effect (Schneider pharmacokinetic model, maximal flow rate <700mL/h) at an initial concentration (Ce) of 4.0 μg/mL.

MCI for bidirectional endoscopy

The MCI group were inducted by a bolus propofol dose of 1.5mg/kg with the same rate limitation (<700mL/h) followed by a maintenance dose at 6–12mg/kg/h as previously reported

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation

I: 2.93 ± 2.61 min

C: 2.43 ± 1.78 min

(p=0.091)

Propofol consumption

I: 32.02 ± 12.09 mL

C: 34.22 ± 9.19 mL

(p=0.119)

Absolute mean difference: -2.20 mL [95%CI -5.04; 0.64]

Adverse events

Mild oxygen desaturation (SpO₂ <90%)

I: 34 (30.9%)

C: 49 (44.5%)

RR 0.694 (95%CI 0.490-0.983)

(p=0.037)

Moderate oxygen desaturation (SpO₂ <85%)

I: 17 (15.5%)

C: 34 (30.9%)

RR 0.50 (95%CI 0.298-0.840)

(p=0.007)

Hypotension

Not reported

Patient satisfaction (comfortable)

I: 7 (7-7)

C: 7 (7-7)

(p=0.276)

Patient satisfaction (free of nausea)

I: 7 (7-7)

C: 7 (7-7)

(p=0.407)

Patient satisfaction (same sedative quality next time)

I: 7 (7-7)

C: 7 (7-7)

(p=0.090)

Author conclusions: the current study demonstrated that TCI using propofol with alfentanil presented faster recovery, better hemodynamic and respiratory stability, and fewer adverse events than MCI in same-day bidirectional endoscopy with targeted deep sedation. In addition, patients in the MCI group had more concerns

and required more assistance to maintain upper airway patency during this bidirectional endoscopic procedure.

De Vito, 2017

Type of study: RCT

Setting and country: Italy

Funding and conflicts of interest: No reporting on funding; no conflict of interest

Inclusion criteria: obstructive sleep apnea (OSA) patients, Apnea-Hypopnea Index (AHI) of 15-30, men and women, age 18-65 years, BMI <30, awake oxygen saturation >85% and able to read and sign consent form.

Exclusion criteria: Chronic Obstructive Pulmonary Disease (COPD), liver disease (Child Pugh 1–3), history of chronic use of sedatives, narcotics, alcohol or

illicit drugs, history of 1st and 2nd degree heart block (not paced), left ventricular Ejection Fraction (EF) <50, allergy to propofol or to any medications, OSHAS surgical failure patients and pregnant women.

N total at baseline:

Intervention: 123

Control: 50

Important prognostic factors²:

age ± SD:

I: 51 ± 11 years

C: 48 ± 12 years

Sex:

I: 91% M

C: 86% M

Sleep endoscopy was performed by a TCI system using Schnider model in effect-site (cerebral) targeted infusion 50-ml prefilled syringe of 1% propofol.

The starting dose of propofol was 1.5 mcg ml⁻¹ and increments of 0.2 mcg ml⁻¹ took place when the new cerebral concentration of propofol was reached. In this way, we realized a slow technique of TCI propofol infusion.

A sleep endoscopy was manually performed using 1 mg kg⁻¹ of propofol as induction bolus and 20 mg of boluses was repeated every 2 min, with a 20-ml syringe.

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation

Not reported

Propofol consumption

I: 148 ± 37 mg

C: 132 ± 32 mg

(p<0.05)

Absolute mean difference: 16.00 mg [95%CI 4.98; 27.02]

Adverse events

Severe oxygen desaturation

I: 0 (0%)

C: 8 (16%)

(p<0.05)

Hypotension

I: 0 (0%)

C: 0 (0%)

Patient satisfaction

Not reported

Author conclusions: Even

in experienced hands, conventional DISE can potentially be ineffective and dangerous. Our results suggest that the DISE-TCI technique should be the first choice in performing sleep-endoscopy because of its increased accuracy, stability and safety, even though a clear polysomnografic analysis is imperative to allow us to differentiate OSA patients in whom UA anatomical abnormalities are predominant in comparison with not-anatomical pathophysiologic factors.

Randomization was not 1:1, intervention was performed up to jan 2015, while control was performed up to jan 2011. Unclear how this study was performed exactly.

Franzen, 2016

Type of study: RCT (noninferiority)

Setting and country: Switzerland

Funding and conflicts of interest: No reporting of funding or conflict of interest

Inclusion criteria: patients undergoing flexible bronchoscopy

Exclusion criteria: Any contraindication to use propofol for sedation; Body mass index > 35 kg/m2; Immunosuppressive therapy after organ transplantation; Antiretroviral therapy for HIV infection; Known or suspected drug or alcohol abuse; Inability to follow the procedures of the study ; Participation in another study with either an investigational drug within 30 days preceding and during the present study; Previous enrolment into the current study; Enrolment of the investigator, his/her family members, employees and other dependent persons; bronchoscopy in attendance of an anaesthesiologist; Age below 18 or above 85 years.

N total at baseline:

Intervention: 39 patients

Control: 38 patients

Important prognostic factors²:

age ± SD:

I: 65.6 ± 11.3 yrs

C: 64.0 ± 11.6 yrs

Sex:

I: 63% M

C: 59% M

TCI for flexible bronchoscopy

TCI of propofol.

Bronchoscopy was started after reaching the initial targeted effect site concentration (Ce) of 2.5 μg/mL propofol using the Schnider pharmacokinetic model described elsewhere, which is based on certain biometric values as total weight, lean body mass and

height.6,13–16 Thereafter, Ce could be adjusted depending on the clinical effect by increments of 0.2 μg/mL, in order to maintain the required level of sedation

Fractionated propofol administration for flexible bronchoscopy

Fractionated propofol administration using non-anaesthesiologist administration of propofol.

The loading doses of propofol were titrated in order to achieve the adequate sedation level (onset of ptosis). Patients received an initial bolus of 30–40mg of propofol, followed by a carefully titrated dose of 10–20mg of propofol based on the clinical response. Between each bolus, a pause lasting more than 20 s had to be observed.

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation

Not reported

Propofol dose

I:  254.1 ± 105.8 mg

C: 193.7 ± 95.0 mg

(p=0.003)

Absolute mean difference: 60.40 mg [95%CI 15.51; 105.29]

Adverse events

Oxygen desaturation (SpO2 <90%) (mean number events/min)

I: 0.13 ± 0.10

C: 0.17 ± 0.14

(p=0.35)

Hypotension (SBP <90 mmHg) (mean number events/min

I: 0.02 ± 0.05

C: 0.01 ± 0.03

(p=0.27)

Patient satisfaction

Not reported

Author conclusions:

In conclusion, we suggest that TCI of propofol is a favourable sedation technique for FB. Safety issues (oxygenation, ventilation and blood pressure), cough frequency and procedure times are comparable with FPA. Using TCI, fewer interventions are needed to induce and maintain sedation. However, the cumulative dose of propofol is higher with the use of TCI, which is maybe negligible, particularly when Ce is adapted to different stages of FB.

A per-protocol analysis was performed were about 20% of patients were excluded. Patients who received a maximum of more than 10 L/min of oxygen or an intervention via a nasopharyngeal airway insertion during the procedure were excluded from this analysis.

Sakaguchi, 2011

Type of study: RCT

Setting and country: Japan

Funding and conflicts of interest:

Funded by non-commercial grant, no statement on conflict of interest

Inclusion criteria: age 16 years or older; ASA physical status 1 or 2; treatment for dental caries, endodontics, periodontics, and prosthesis; moderate or severe intellectual disability

Exclusion criteria: a cooperating patient who could communicate; premedication; uncontrolled or severe medical condition

N total at baseline:

Intervention: 20 patients

Control: 20 patients

Important prognostic factors²:

Age ± SD:

I: 30.5 ± 11.2

C: 30.5 ± 10.8

Sex:

I: 60% M

C: 75% M

Bispectral index (BIS)-guided TCI for dental treatment

Propofol was infused with a TCI pump at a target concentration of 1.5 µg/ml.

Manually controlled infusion of propofol without BIS monitoring for dental treatment

Several bolus injections of 10 mg propofol were given at 20-second intervals until patients achieved assessment of Behavior Reactions (ABR) score 2. Thereafter propofol was continuously infused at a rate of 4.0 mg/kg/r, using a TCI pump in manual infusion mode.

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation

Not reported

Average propofol consumption (mean ± SD):

I: 76.1 ± 18.8 µg/kg/min

C: 93.3 ± 18.8 µg/kg/min

(p=0.0061)

Absolute mean difference: -17.20 µg/kg/min [95%CI -28.85; -5.55]

Adverse events

Hypotension

I: 0 (0%)

C: 0 (0%)

Oxygen deprivation

I: 0 (0%)

C: 0 (0%)

Patient satisfaction

Not reported

Author conclusions: In conclusion, BIS-guided TCI over conventional manual control 1) reduced the dose requirement of propofol and 2) promoted quicker recovery from sedation.

Authors state that there were no significant complications.

Vučićević, 2016

Type of study: RCT

Setting and country: Serbia

Funding and conflicts of interest: No funding or conflict of interest reported

Inclusion criteria: patients of both sexes, 18-65 years old, classified into group 1 or 2 of ASA scheduled for diagnostic outpatient colonoscopy with deep sedation.

Exclusion criteria: those allergic to the study drugs, patients with previous problems with anesthesia or sedation, patients with history of stridor, snoring or sleep apnea, patients with neck abnormalities, and those classified into groups III or IV of Mallampati classification [12], patient with neuropsychiatric, cardiac, respiratory, renal disorders, those in pregnant state, and with history of large-bowel surgery.

N total at baseline:

Intervention: 45

Control: 45

Important prognostic factors²:

age ± SD:

I: 50.35 ± 11.21 years

C: 49.02 ± 11.91 years

Sex:

I: 33% M

C: 42% M

TCI propofol for colonoscopy

Patients received propofol with TCI pump according to the Schnider’s pharmacokinetic model, with the initial Ce of 2.5 μg/ml. This concentration was increased or decreased by 0.5–1 μg/ml until the deep level of sedation was achieved.

Manual intravenous titration technique of propofol for colonoscopy

Patients received propofol intravenously (10 mg/ml), in a bolus of 0.5 mg/kg, and then 10–20 mg were titrated every one to two minutes.

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation (induction time)

I: 3.28 ± 0.62 min

C: 3.31 ± 0.55 min

(p=0.859)

Propofol consumption

I: 148.97 ± 58.46 mg

C: 155.77 ± 52.63 mg

(p=0.563)

Absolute mean difference: -6.80 mg [95%CI -29.78; 16.18]

Adverse events

Severe oxygen desaturation

Not reported

Hypotension

Not reported

Patient satisfaction

Not reported

Oxygen and blood pressure levels were reported, however not above/under certain thresholds. This was stated in the methods, but was not reported in the results.

Wang, 2016

Type of study: RCT

Setting and country: China

Funding and conflicts of interest: non-commercial funding from Changhai Hospital; no conflict of interest

Inclusion criteria: patients meeting indication of elective colonoscopy, age >18 years old, ASA status 1 or 2

Exclusion criteria: morbid obesity (BMI ≥30 kg/m2); predicted difficult airway; hypertension; known cardiac diseases; acute or chronic hepatic and renal dysfunction; long-term use of anesthetics or opioids; and patients undergoing invasive procedures under colonoscopy.

N total at baseline:

Intervention: 36 patients

Control: 36 patients

Important prognostic factors²:

age ± SD:

I: 42.6 ± 7.6 yrs

C: 40.2 ± 6.9 yrs

Sex:

I: 58% M

C: 64% M

TCI propofol for colonoscopy

In the TCI group, propofol was infused through the Module DPS TCI system using the Marsh model. The primary plasma concentration was set as 3.0 μg/ml and an adjustment of 0.2 μg/ml was made upon the patients’ response based on the Observer’s Assessment of Alertness/Sedation (OAAS) Score.

Manually-controlled infusion (MCI) propofol for colonoscopy.

In the MCI group, propofol was injected at a bolus of 1.5 mg/kg and then continuously infused at 6 mg/kg/h using the conventional continuous microinfusion pump. A bolus of 0.5 mg/kg propofol

was be injected as required.

All outcomes are measured during or on the same day as the procedure

Outcome measures and effect size (include 95%CI and p-value if available):

Effectivity

Time to sedation

Not reported

Propofol dose

Not reported

Adverse events

Oxygen desaturation (SpO2 <90%)

I: 0 (0%)

C: 0 (0%)

Hypotension (mean blood pressure <55 mmHg)

I: 0 (0%)

C: 0 (0%)

Patient satisfaction

I: 77.2 ± 6.7

C: 76.8 ± 7.8

(p=0.808)

Author conclusions:

Given the fact that training residents of anesthesiology were less experienced in colonoscopy sedation, propofol TCI is an ideal infusion mode for them to improve sedation quality and promote stable cardiovascular and respiratory status.

Cross over was performed on the 18 training residents in anaesthesiology, they first received TCI or MCI training for one month, and then vice versa for the second month. The last two patients of each month of each training resident were used in this study.

Study reference

Was the allocation sequence adequately generated? ^a

Definitely ye

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?^b

Definitely yes

Probably yes

Probably no

Definitely no

Blinding: Was knowledge of the allocated

interventions adequately prevented?^c

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?^d

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?^e

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?^f

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure^g

LOW

Some concerns

HIGH

Chang, 2015

Definitely yes

Reason: Computer-generated randomization blocks.

No information

Reason: -

Definitely no

Reason: Health care providers not blinded, blinding of patients, outcome assessors blinded, data collectors and analysts not reported.

Definitely yes

Reason: All outcomes were measured during or on the same day as the procedure

Definitely yes

Reason: All relevant outcomes were reported

Definitely yes

Reason: No other problems noted

Some concerns (propofol consumption, adverse events)

Reason: potential problems with concealment of randomization and blinding.

Chiang, 2013

Definitely yes

Reason: computer-generated number table for a single-factor complete randomization

Definitely yes

Reason: randomization performed by independent data manager not involved in study analysis. Opaque envelops were used and opened just before each procedure.

Probably yes

Reason: patients, health care providers and outcome assessors were blinded. Anaesthesiologists were not blinded, but their infusion protocols were controlled.

Definitely yes

Reason: All outcomes were measured during or on the same day as the procedure

Definitely yes

Reason: All relevant outcomes were reported

Definitely yes

Reason: No other problems noted

LOW (propofol consumption, adverse events, patient satisfaction)

De Vito, 2017

Definitely yes

Reason: Computer-generated schedule for random allocation

Probably yes

Reason: Sealed in numbered envelopes

Probably no

Reason: Blinding was not reported, but a blinded revision was conducted on reported outcomes

Definitely yes

Reason: All outcomes were measured during or on the same day as the procedure

Definitely yes

Reason: All relevant outcomes were reported

Probably no

Reason: there was no information on the comparability of the study groups at baseline.

HIGH (propofol consumption, adverse events)

Reason: potential problems with blinding, no information on comparability of study groups.

Franzen, 2016

Definitely yes

Reason: Randomized using sequentially numbered sealed opaque enveloped

No information

Reason: -

Probably no

Reason: Patients blinded, health care providers and outcome assessors not blinded (blinding of data collectors and data analysis not reported.)

Definitely yes

Reason: All outcomes were measured during or on the same day as the procedure

Definitely yes

Reason: All relevant outcomes were reported

Definitely yes

Reason: No other problems noted

LOW (propofol consumption, adverse events)

Sakaguchi, 2011

Definitely no

Reason: Allocation based on clinical record number

Probably no

Reason: Allocation performed by non-investigator at hospital registration

Probably no

Reason: Health care providers and outcome assessors blinded, anestesiologist not blinded, patients were probably not blinded (blinding of data collectors and analysts not reported)

Definitely yes

Reason: All outcomes were measured during or on the same day as the procedure

Definitely yes

Reason: All relevant outcomes were reported

Definitely yes

Reason: No other problems noted

Some concerns (propofol consumption, adverse events)

Reason: potential problems with randomization and blinding.

Vučićević, 2016

Probably yes

Reason: randomization based on random-numbers table

No information

Reason: -

Probably no

Reason: Blinding was not reported

Definitely yes;

Reason: All outcomes were measured during or on the same day as the procedure

Definitely no

Reason: Adverse events were not reported in result section.

Definitely yes;

Reason: No other problems noted

Some concerns (time to sedation, propofol consumption)

Reason: no information on allocation concealement, blinding was not reported

HIGH (adverse events)

Reason: Adverse events were not reported in result section, while they were mentioned as outcomes.

Wang, 2016

Probably no

Reason: No information on randomization of patients, only randomization of anaesthesiology residents (randomly allocated based on computer-generated numbers)

No information

Reason: -

Probably no

Reason: Endoscopists and patients were blinded, anesthesiologists, outcome assessors were not blinded. (blinding of data collectors and analysts not reported)

Definitely yes;

Reason: All outcomes were measured during or on the same day as the procedure

Definitely yes;

Reason: All relevant outcomes were reported

Definitely yes;

Reason: No other problems noted

HIGH (adverse events, patient satisfaction)

Reason: patients were not randomized, there was no information on allocation concealment and blinding of outcome assessors was not performed.