doi:10.1111/codi.12837

Systematic review

Carbon dioxide insufflation vs conventional air insufflation for colonoscopy: a systematic review and meta-analysis of published randomized controlled trials M. S. Sajid*, J. Caswell*, M. I. Bhatti†, P. Sains*, M. K. Baig* and W. F. A. Miles† *Department of General, Endoscopic and Laparoscopic Colorectal Surgery, Western Sussex Hospitals NHS Trust, Worthing Hospital, Worthing, West Sussex and †Department of General and Colorectal Surgery, Queen Elizabeth Hospital, King’s Lynn NHS Foundation Trust, King’s Lynn, Norfolk, UK Received 7 June 2014; accepted 6 August 2014; Accepted Article online 12 November 2014

Abstract Aim Conventional air insufflation (AI) may cause prolonged abdominal bloating, excessive abdominal pain and discomfort during colonoscopy. Carbon dioxide may be an acceptable alternative to avoid these complications. The object of this study was to evaluate systematically the effectiveness of carbon dioxide insufflation (CI) for colonoscopy compared with AI. Method Randomized controlled trials (RCTs) comparing the effectiveness of CI with that of AI during colonoscopy were retrieved from medical electronic databases and combined analysis was performed using the REVMAN statistical package. The combined outcome of dichotomous and continuous variables was expressed as an odds ratio (OR) and standardized mean difference (SMD). Results Twenty-one RCTs comprising 3607 patients were included in the study. There was statistically signif-

Introduction Colonoscopy is the preferred investigation for colorectal cancer and the less invasive mode of treatment for colorectal polyps [1,2]. As many as 30–60% of patients report abdominal pain, bloating and discomfort following colonoscopy [3]. When colonoscopy is performed for screening it is imperative that efforts are made to minimize the colonoscopy-related abdominal pain and Correspondence to: Muhammad Shafique Sajid MBBS, MSc, MBA, FRCS, Surgical Specialist Registrar, Washington Suite, North Wing, Worthing Hospital, West Sussex BN11 2DH, UK. E-mail: [email protected] The provisional abstract of this study has been presented orally at the 101st Annual Congress of the Association of Surgeons of Great Britain and Ireland in Harrogate on 30 April to 2 May 2014. The abstract will be published in the Supplement issue of the British Journal of Surgery.

icant heterogeneity among included studies. CI showed a significant trend towards reduced procedural pain [SMD –1.34; 95% confidence interval (95% CI) 2.23 to 0.45; z = 2.96; P < 0.003] and also postprocedural pain at 1 h (SMD –1.11; 95% CI 1.83 to 0.38; z = 2.97; P < 0.003), 6 and 24 h (OR 0.44; 95% CI 0.23–0.85; z = 2.44; P < 0.01). CI was associated with faster caecal intubation (SMD 0.20; 95% CI 0.37 to 0.02; z = 2.23; P < 0.03) but the caecal intubation rate was similar (P = 0.59) in both colonic insufflation techniques . Conclusion CI seems to have clinical advantages over AI for colonoscopy with regard to pain during and after the procedure. Keywords Carbon dioxide, insufflation, colonoscopy, air, pain, distension

distension that is mainly attributed to air insufflation (AI) [4]. However, AI is still the most often used method for colonic insufflation during colonoscopy. Sedative and analgesic protocols have been established to reduce pain and procedural stress [5,6]. In some patients colonoscopy without sedation may save up to an hour in recovery time and nursing care [7]. Hence, minimizing pain in unsedated patients is a growing priority. Antispasmodic agents have been widely used [8,9], but a recent study has reported a reduced adenoma detection rate with their use [10]. Technical details such as patient position, water immersion, normal and warm water insufflation have been reported [11–15] with variable results in terms of the degree of colonoscopy-induced abdominal pain and the learning curve for the endoscopist. Carbon dioxide

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

111

CO2 for colonoscopy

Identification

M. S. Sajid et al.

Records identified through database search (n = 77)

Additional records identified through other sources (n = 0)

Included

Eligibility

Screening

Records after duplicates removed (n = 74)

Records screened (n = 74)

Full-text articles assessed for eligibility (n = 42)

Records excluded (n = 32) Reason Irrelevant

Full-text articles excluded, (n = 21) Reasons Other reviews Other technique reviews Duplicate data

Studies included in the meta-analysis (n = 21)

Figure 1 PRISMA flow diagram.

(CO2) insufflation (CI) has shown promising results, suggesting this could be a means of reducing pain. CI was first suggested in 1953 as an alternative to reduce the risk of gas combustion during electrocoagulation [16]. Further studies demonstrated that CO2 was absorbed from the colon 150 times faster than nitrogen and reduced the risk of the bowel ischaemia [17,18], resulting in reduced spasm and pain [4]. The aim of the present study was to evaluate systematically the effectiveness of CI for colonoscopy compared with AI.

insufflation’, ‘CO2 inflation’ and ‘carbon dioxide insufflation’. No limits for language, gender, sample size and place of study origin were entered for the search. Boolean operators (AND, OR, NOT) were used to narrow and widen the search results. The titles from the search results were determined as suitable for potential inclusion in the study. In addition, the references from selected articles were examined as a further search tool to find additional trials.

Method

Study selection

Data sources and search strategy

We searched for relevant articles for this review using electronic databases such as MEDLINE and EMBASE and the Cochrane Library for randomized, controlled trials (RCTs). The Medical Subject Headings (MeSH) search terms published in the MEDLINE library were used to identify relevant RCTs. The terms ‘colonoscopy’, ‘gastrointestinal endoscopy’, ‘lower gastrointestinal endoscopy’ and ‘endoscopy’ were combined with the MeSH terms ‘air insufflation’, ‘air inflation’, ‘CO2

112

For inclusion in the meta-analysis a study had to meet the following criteria: (i) RCT, (ii) comparison between CI and AI, (iii) evaluation of patient experience of pain, discomfort, caecal intubation time and caecal intubation failure rate, and (iv) main outcome measures reported preferably as an intention-to-treat (ITT) analysis. Data extraction

Two independent reviewers using a predefined metaanalysis form extracted data from each study. This

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

CO2 for colonoscopy

M. S. Sajid et al.

Table 1 Characteristics of the included studies. Gender M: F

Trial

Year

Country

Age (years)

Amato et al. [29]

2013

Italy

Bretthauer et al. [30]

2002

Norway

Bretthauer et al. [31]

2003

Norway

61.5  14 (CI) 60  13.4 (AI) 59.5 (CI) 59.6 (AI) 59 (55–64)

Bretthauer et al. [32]

2005

Norway

50 (22–94)

Chen et al. [33]

2013

Taiwan

Church et al. [34]

2003

USA

55.47  12.36 (CI) 55.75  13.55 (AI) No data

Cleland et al. [35]

2013

New Zealand

Diez-Redondo et al. [36] Geyer et al. [37]

2012

Spain

2011

Switzerland

Iida et al. [38]

2013

Japan

Imai et al. [39]

2012

Japan

Liu et al. [40]

2009

China

Mayr et al. [41]

2012

Germany

Riss et al. [42]

2009

Austria

Seo et al. [43]

2013

Korea

Singh et al. [44]

2012

Australia

Stevenson et al. [45] Sumanac et al. [46]

1992

Canada

2002

Canada

Uraoka et al. [47]

2009

Japan

Wong et al. [48]

2008

China

Yamano et al. [49]

2010

Japan

Follow-up (h)

Indications for colonoscopy

24 24

Screening, surveillance, diagnostic Screening

24

Screening

24

Diagnostic

24

Screening, surveillance, diagnostic Screening, surveillance, diagnostic Diagnostic

61.33  15.79 (CI) 61.93  12.75 (AI) 56 (24–82) (CI) 56.7 (24–83) (AI) 58  13 (CI) 62  12 (AI) 58.9  10.8 (CI) 56.6  12.2 (AI) 55.7  18.6 (CI) 57.8  17.7 (AI) 48.5  16.8 (CI) 48.1  13.2 (AI) 62.4  8.3 (CI) 61.0  10.9 (AI) Not reported

75:40 (CI) 73:40 (AI) 77:44 (CI) 75:44 (AI) 68:41 (CI) 67:42 (AI) 26:26 (CI) 26:25 (AI) 56:40 (CI) 54:43 (AI) 54:69 (CI) 73:51 (AI) 51:57 (CI) 43:54 (AI) 59:70 (CI) 63:78 (AI) 42:68 (CI) 58:51 (AI) 36:11 (CI) 38:12 (AI) 9:10 (CI) 8:10 (AI) 100:74 (CI) 101:74 (AI) 35:42 (CI) 32:47 (AI) Not reported

48.8  9.0 (CI) 49.9  8.4 (AI) 58.26 (CI) 59.97 (AI) Not reported

20:28 (CI) 24:22 (AI) 45:25 (CI) 33:39 (AI) Not reported

24

55.9  1.9 (CI) 53.1  1.5 (AI) 65 (59–73) (CI) 62 (47–71) (AI) 59.9  15.2 (CI) 58.5  12.3 (AI) 63.2  8.5 (CI) 61.7  9.3 (AI)

Equal distribution 18:39 (CI) 19:38 (AI) 21:23 (CI) 23:26 (AI) 41:25 (CI) 37:17 (AI)

24

resulted in high interobserver agreement based on the results of the Kappa statistic (0.91). Information related to trial methodology and trials results were collected and secured on an encrypted datasheet. After completing the data extraction the two independent reviewers discussed the results, and any discrepancies were resolved by consensus.

1 1 24 24 1

Screening, surveillance, diagnostic Screening, surveillance Diagnostic

24

Diagnostic, screening

24

Not reported

24

Diagnostic, screening

12

Screening, surveillance, diagnostic Screening, surveillance, diagnostic Screening, surveillance

~3 24

6 2 24

Sequential colonoscopies for various indications Screening, diagnostic Screening, surveillance, diagnostic Diagnostic Screening

Statistical analysis

The software package REVMAN 5.3 [19,20], provided by the Cochrane Collaboration, was used for the statistical analysis. The odds ratio (OR) with a 95% confidence interval (95% CI) was calculated for binary data and the standardized mean difference (SMD) with a 95% CI was

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

113

CO2 for colonoscopy

M. S. Sajid et al.

Table 2 Treatment protocol adopted in the included trials. Trial

CO2 insufflation group

Air insufflation group

Amato et al. [29]

Sedation and analgesia given on patient request only Procedure by an experienced gastroenterologist Olympus HD180 series colonoscope was used Olympus UCR CO2 intraluminal insufflation unit was used for CO2 insufflation Bowel preparation formula was not reported Procedure by an experienced endoscopist No routine use of sedation Bowel cleansing with 4 l polyethylene glycol solution Olympus video colonoscope CO2 or air administered using two different pressures and flow-controlled devices connected to the CO2 and air reservoirs provided: endoscopic CO2 regulator Key Med Ltd for CO2 As per above As per above 2 l polyethylene glycol 6–9 h before colonoscopy Two experienced endoscopists performed the procedure Standard video colonoscopes (Olympus CF H260AZI) Routine sedation was not used Left lateral position was switched to supine after sigmoid colon was passed Olympus UCR endoscopic CO2 regulation unit gas insufflator was sued for CI Two experienced endoscopists performed the procedure Bowel preparation with either polyethylene glycol or sodium phosphate Selective meperidine for analgesia with midazolam Olympus 140 and 160 series colonoscopes Regulator set to deliver CO at 3 l/min Olympus CO2 delivery system for CI Flow meter set at 4 l/min Endoscopy was performed by endoscopist of variable experience Hypnovel and fentanyl were routine sedatives and the use of buscopan was selective Endoscopy by six experienced endoscopists and a trainee Under sedation of midazolam and propofol Colonic preparation was not specified Colonoscopes Exera II CFH180 AL. UCR Olympus CO2 insufflator All colonoscopies completed by experienced endoscopist Standard Pentax endoscopes with standard processor CO2 efficient insufflator device (Westbury)

As per CI group except AI was used AI colonoscopy performed in standard fashion. Aliquots of 30–50 ml water at room temperature allowed for washing residual stool as required

Bretthauer et al. [30]

Bretthauer et al. [31] Bretthauer et al. [32] Chen et al. [33]

Church et al. [34]

Cleland et al. [35]

Diez-Redondo et al. [36]

Geyer et al. [37]

114

Norsk Hydro Ltd endoscopic regulator for AI

As per CI except AI was used As per CI except AI was used As per CI group but CO2 was used at the time of extubation

As per CI except AI was used

As per CI except AI was used

As per CI except AI was used

As per CI except AI was used

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

CO2 for colonoscopy

M. S. Sajid et al.

Table 2 (Continued). Trial

Iida et al. [38]

Imai et al. [39]

Liu et al. [40] Mayr et al. [41]

Riss et al. [42]

Seo et al. [43]

Singh et al. [44]

Stevenson et al. [45]

CO2 insufflation group Basal flow rate 0.5 l (up to 3 l/min) All patients sedated with propofol Day before the procedure all patients received 24 mg sennoside at 9 p.m. and on the day of procedure 2000 ml of polyethylene glycol All patients received 10 mg of scopolamine butylbromide as antispasmodic No sedatives or analgesia Three experienced blinded endoscopists performed the procedure CO2 Olympus UCR regulator Olympus CF 240 or CF Q260AI colonoscopes Dietary restrictions before procedure Olympus UCR CO2 regulator Olympus Q-240I electric colonoscopes One experienced endoscopist 10 mg scopolamine butylbromide or 1 mg glucagon were used as antispasmodic Castor oil was used for bowel preparation Further information was not reported UCR endoscopic CO2 regulator was used One endoscopist with vast experience performed the procedure No further information were reported Dietary restrictions before procedure Full bowel prep with polyethylene glycol based antegrade gut lavage Procedure was performed by an experienced endoscopist Propofol and midazolam for moderate sedation Olympus video endoscope and commercial CO2 insufflation system (Intramed) Dietary restrictions before procedure 2 l polyethylene glycol at 6 p.m. on day before procedure and remaining 2 l at 5–7 a.m. at least 2 h before procedure. Patients instructed to drink 250 ml of polyethylene glycol every 10 min Three experienced endoscopists performed the procedure Pethidine and midazolam was used for sedation Flumazenil was given postprocedure Colosence CO2 insufflator CO2 was delivered at rate of 4 l/min with 50 kPa pressure using standard CO2 delivery system Experienced gastroenterologists or colorectal surgeons performed colonoscopy Propofol, midazolam or fentanyl was sued for sedation 100 mg meperidine and 10 mg diazemuls used for sedation One experienced endoscopist performed the colonoscopy

Air insufflation group

As per CI except AI was used

As per CI except AI was used

As per CI except AI was used As per CI except AI was used

As per CI except AI was used

As per CI except AI was used

As per CI except AI was used

As per CI except AI was used

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

115

CO2 for colonoscopy

M. S. Sajid et al.

Table 2 (Continued). Trial

CO2 insufflation group

Air insufflation group

Sumanac et al. [46]

Olympus ECR delivery system was used for CO2 insufflation Diazemuls and merperidine was used for sedation Sodium picosulphate and 2 l polyethylene glycol used for bowel preparation Scopolamine butylbromide or glucagon to suppress bowel movement Selective use of midazolam Olympus variable-stiffness colonoscope was used CO2 commercial regulator connected to CO2 bottle at rate of 2 l/min Eight colonoscopists performed the procedure Olympus video-colonoscopes and insufflator CO2 insufflator also from Olympus Two experienced endoscopists performed procedure Diazemuls and pethidine for sedation were used Two models of video-colonoscopes were used Olympus UCR endoscopic CO2 regulation unit was used for both CI and AI No sedation All patients received antispasmodic scopolamine butylbromide as antispasmodic

As per CI except AI was used

Uraok et al. [47]

Wong et al. [48]

Yamano et al. [49]

calculated for continuous data variables. The randomeffects model [21,22] was used to calculate the combined outcomes. Heterogeneity was explored using the chi-square test, with significance set at P < 0.05, and was quantified using I2 [23], with a maximum value of 30% identifying low heterogeneity [23]. The Mantel– Haenszel method was used for the calculation of the OR under the random effect model [24,25]. If the standard deviation was not available then it was calculated according to the guidelines of the Cochrane Collaboration [21]. A forest plot was used for the graphical display of the results. The methodological quality of the included trials was initially assessed using the published guidelines of Jaddad et al. [26] and Chalmers et al. [27]. Based on the quality of the included RCTs, the strength and summary of the GRADE quality of evidence was achieved using GradePro [28], a tool provided by the Cochrane Collaboration. End-points

The primary end-point was the mean pain score and the total number of patients experiencing pain during the procedure and 1, 6 and 24 h after the procedure. Secondary end-points included the caecal intubation failure rate and time to arrive at the caecum.

116

As per CI except AI was used

As per CI except AI was used

As per CI except AI was used

Results The PRISMA flow chart is given in Fig. 1. Twenty-one RCTs [29–49] on 3607 patients were retrieved from the electronic databases. There were 1801 patients in the CI group and 1806 in the AI group (Table 1). The salient features and treatment protocols adopted in the included trials are given in Table 2. Methodological quality of included RCTs

Based on the guidelines of Jaddad et al. [26] and Chalmers et al. [27], the quality of the majority of included trials was good to moderate, largely owing to the fact that randomization was used. There was adequate reporting of the power calculation and allocation concealment (Table 3). However, most trials did not specify whether ITT was used. Based on the quality of the RCTs, the GRADE quality of evidence is shown in Fig. 2. Pain during colonoscopy

Nine trials included data on pain (Fig. 3a). There was significant heterogeneity (s2 = 40.0, v2 = 28.3, d.f. = 8, P = 0.0005; I2 = 71%) among them. In the

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

CO2 for colonoscopy

M. S. Sajid et al.

Table 3 Quality parameters of the included trials. Power calculations

Blinding

Concealment

ITT

Computer-generated random number Use of blinded envelopes

Yes

Yes for patient

Yes

Yes

Yes

Yes

No

Single day sessions randomized

No

Yes

No

No

Yes

No

Yes

Yes

No

No

Yes

No

Diez-Redondo et al. [36] Geyer et al. [37]

Sealed envelopes to randomly assign the patients Computer-generated randomization Randomized to Room 1 (CI) and Room 2 (AI) Sealed envelopes to randomly assign the patients Computer-generated randomization Dice used for randomization

No

No

Yes

No

Iida et al. [38]

Technique was not reported

No

No

Imai et al. [39]

Random number allocation

Yes

No

Liu et al. [40]

Random selection by an endoscopy nurse Computer-generated randomization Computer-generated randomization Computer-generated randomization Sealed envelopes Randomized number technique

Yes for patient Yes for endoscopist Endoscopist and participants both blinded to type and volumes of gas used Yes for patient Yes for endoscopist Endoscopist, patients as well the research staff were blinded Endoscopist, patients as well the research staff were blinded Endoscopist, patients as well the research staff were blinded Patient, nurse and endoscopist were blinded Patient and endoscopist were blinded Patient and endoscopist were blinded Patient and endoscopist were blinded Single blinded

Yes

No

Patient and endoscopist were blinded Only patients were blinded

Yes

No

No

Yes

Patient, endoscopist and recovery nurses were blinded Patient and endoscopist were blinded Patient and endoscopist were blinded Patient and endoscopist were blinded Patient and endoscopist were blinded Endoscopist, patients as well the research staff were blinded Patient and endoscopist were blinded

Yes

Yes

Yes Yes

No No

Yes

Yes

Yes

Yes

Yes

No

No

No

Trial

Randomization

Amato et al. [29] Bretthauer et al. [30] Bretthauer et al. [31] Bretthauer et al. [32] Chen et al. [33] Church et al. [34] Cleland et al. [35]

Mayr et al. [41] Riss et al. [42] Seo et al. [43] Singh et al. [44] Stevenson et al. [45] Sumanac et al. [46]

Yes No Yes Yes Yes No

No Yes Yes Yes No No

Random number list

No

Uraok et al. [47]

Consecutive patient numbers

Yes

Wong et al. [48]

Sealed envelopes

Yes

Yamano et al. [49]

Technique was not reported

No

random effects model the risk of having pain during the procedure was higher following AI (OR 0.50; 95% CI 0.30–0.84; z = 2.64; P < 0.008). Ten trials contributed to the combined calculation of the pain score (Fig. 3b). There was significant heterogeneity among trials (s2 = 1.24, v2 = 355.98, d.f. = 9, P = 0.00001; 2 I = 97%). In the random effects model CI was

associated with a reduced pain score (SMD –1.35; 95% CI –2.05 to –0.64; z = 3.76; P < 0.0002). Pain 1 h after colonoscopy

Seven included trials had data on pain immediately after colonoscopy (Fig. 4a). There was significant heteroge-

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

117

CO2 for colonoscopy

M. S. Sajid et al.

CO2 insufflation for colonoscopy Patient or population: patients with colonoscopy Settings: Intervention: CO2 insufflation Outcomes

Pain during procedure in number of patients Odds ratio Follow-up:mean 1 days

Pain score during procedure Standardised mean difference Follow-up: mean 1 days Pain after 1 hour in number of patients Odds ratio Follow-up: mean 1 days

Pani score after 1 hour Standardised mean difference Follow-up: mean 1 days Pain after 6 hours in number of patients Odds ratio Follow-up: mean 1 days

Pani score after 6 hours Standardised mean difference Follow-up: mean 1 days Pain after 24 hours in number of patients Odds ratio Follow-up: mean 1 days

Relative effect No of participants Quality of the evidence Comments (studies) (GRADE) (95% Cl)

Illustrative comparative risks* (95% Cl) Assumed risk Corresponding risk Control CO2 insufflation Study population 531 per 1000 362 per 1000 (254 to 488) Moderate 537 per 1000 367 per 1000 (258 to 493)

1302 (9 studies)

high

1633 (10 studies)

high

1193 (7 studies)

high

1491 (9 studies)

high

OR 0.25 1538 (0.11 to 0.55) (9 studies)

high

1228 (7 studies)

high

OR 0.42 1240 (0.23 to 0.77) (8 studies)

high

390 (2 studies)

high

2295 (12 studies)

high

OR 0.5 (0.3 to 0.84)

The mean pain score during procedure in the intervention groups was

1.35 satndard deviations lower (2.05 to 0.64 lower) Study population 551 per 1000 228 per 1000 (79 to 511) Moderate 451 per 1000 165 per 1000 (54 to 511)

OR 0.24 (0.07 to 0.85)

The mean pain score after 1 hour in the intervention groups was

1.31 satndard deviations lower (2.02 to 0.59 lower) Study population 262 per 1000 82 per 1000 (38 to 163) Moderate 259 per 1000 80 per 1000 (37 to 161) The mean pain score after 6 hours in the intervention groups was

0.58 satndard deviations lower (1.04 to 0.12 lower) Study population 119 per 1000 53 per 1000 (30 to 94) Moderate 114 per 1000 51 per 1000 (29 to 90)

Pani score after 24 hours Standardised mean difference Follow-up: mean 1 days

The mean pain score after 24 hours in the intervention groups was

SMD –1.35 (–2.05 to –0.64)

SMD –1.31 (–2.02 to –0.59)

SMD –0.58 (–1.04 to –0.12)

SMD –0.02 (–0.41 to –0.36)

0.02 satndard deviations lower (0.41 lower to 0.36 higher)

Caecal intubation failure Odds ratio Follow-up: mean 1 days

OR 0.96 (0.63 to 1.45)

Study population 45 per 1000 43 per 1000 (29 to 64) Moderate 21 per 1000 20 per 1000 (13 to 30)

2864 The mean time to caecum in the intervention groups was SMD –0.18 (–0.34 to –0.03) high 0.18 standard deviations lower (15 studies) (0.34 to 0.03 lower) *The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% Cl)

Time of caecum Standardised mean difference Follow-up: mean 1 days

Cl: Confidence interval; OR: Odds ratio; GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

Figure 2 Strength and summary of the evidence analysed on GradePro.

neity among them (s2 = 2.69, v2 = 102.4, d.f. = 6, P = 0.00001; I2 = 94%). In the random effects model the risk of having pain 1 h after colonoscopy was higher following AI. Nine of the trials contributed to the combined calculation of the pain score (Fig. 4b). There was significant heterogeneity among trials (s2 = 1.14, v2 = 294.18, d.f. = 8, P = 0.00001; I2 = 97%). In the random effects model CI was associated with a reduced pain score 1 h after the procedure (SMD 1.31; 95% CI 2.02 to 0.59; z = 3.59; P < 0.0003).

colonoscopy was higher following the use of AI (OR 0.25; 95% CI 0.11–0.55; z = 3.43; P < 0.0006). Seven trials contributed to the combined calculation of pain score (Fig. 5b). There was significant heterogeneity among them (s2 = 0.35, v2 = 86.93, d.f. = 6, P = 0.00001; I2 = 93%). In the random effects model CI was associated with a reduced pain score 6 h after the procedure (SMD 0.58; 95% CI, 1.04, 0.12; z = 2.47; P < 0.01). Pain 24 h after colonoscopy

Pain 6 h after colonoscopy

Nine of the trials contributed to the combined calculation of pain reported by patients (Fig. 5a). There was significant heterogeneity among them (s2 = 0.96, v2 = 32.55, d.f. = 7, P = 0.0001; I2 = 78%). In the random effects model the risk of having pain 6 h after

118

Eight trials contained data on pain 24 h after colonoscopy. There was no significant heterogeneity among them (s2 = 0.17, v2 = 7.31, d.f. = 5, 2 P = 0.20; I = 32%). In the random effects model the risk of having pain 24 h after colonoscopy was higher following the use of AI (OR 0.42; 95% CI 0.23–

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

CO2 for colonoscopy

M. S. Sajid et al.

(a) Study or subgroup Bretthauer 2002 Bretthauer 2005 Church 2003 Cleland 2013 Mayr 2012 Stevenson 1992 Sumanac 2002 Wong 2008 Yamano 2010

Air insufflation Odds ratio CO2 insufflation Events Total Events Total Weight M-H, Random, 95%CI 119 14.6% 0.66 [0.40, 1.11] 71 60 121 43 40 42 5.9% 0.38 [0.07, 2.08] 38 0.77 [0.46, 1.29] 123 51 124 14.6% 43 108 46 97 13.4% 0.21 [0.11, 0.40] 17 79 12.3% 0.34 [0.16, 0.73] 77 28 12 27 21 29 8.3% 1.68 [0.47, 5.95] 22 49 17 51 10.4% 0.33 [0.12, 0.90] 7 44 39 49 8.5% 2.56 [0.74, 8.87] 40 29 54 11.9% 0.19 [0.08, 0.44] 66 12

Total (95% CI)

658

644 100.0%

0.50 [0.30, 0.84]

Total events 251 342 Heterogeneity Tauz = 0.40; Chiz = 28.03, df = 8 (P = 0.0005); Iz = 71% Test for overall effect: Z = 2.64 (P = 0.0008)

(b) Study or subgroup Bretthauer 2002 Chen 2013 Cleland 2013 Diez-Redondo 2012 Geyer 2011 Imai 2013 Singh 2012 Uraoka 2009 Wong 2008 Yamano 2010

Odds ratio M-H, Random, 95% CI

0.1 0.2 0.5 1 Favours CI

CO2 insufflation Air insufflation Std. Mean Difference Mean SD Total Mean SD Total Weight IV, Random,95%CI 2 2 0.43 1.46 1.2 4.5 0.46 2.3 2.34 0.25

0.5 1 1.2 0.5 1.8 0.5 0.9 1.5 0.42 0.5

121 96 108 129 110 19 70 57 44 66

2.75 0.5 3 0.75 1.61 2.35 3.27 0.5 1.2 1.9 4.8 0.5 0.82 0.9 2.6 1.5 4.14 0.4 0.9 0.5

119 97 97 141 109 18 72 57 49 54

10.2% 10.2% 10.2% 10.1% 10.2% 9.5% 10.2% 10.1% 9.3% 10.1%

2 5 10 Favours AI

Std. Mean Difference IV, Random, 95% CI

–1.50 [–1.78, –1.21] –1.13 [–1.43, –0.82] –0.64 [–0.92,–0.36] –3.61 [–4.00, –3.22] 0.00 [–0.26, –0.26] –0.59 [–1.25, –0.07] –0.40 [–0.73, –0.07] –0.20 [–0.57, –0.17] –4.36 [–5.12, –3.60] –1.29 [–1.69, –0.90]

820 813 100.0% –1.35 [–2.05, –0.64] Total(95% CI) Heterogeneity: Tauz = 1.24; Chiz = 355.98, df = 9 (P < 0.00001); Iz = 97% –4 Test for overall effect: Z = 3.76 (P = 0.0002)

–2 0 2 4 Favours CI Favours AI

Figure 3 (a) Forest plot for pain during colonoscopy. Odds ratios are shown with 95% CIs (on plot: CI, CO2 insufflation; AI, air insufflation). (b) Forest plot for pain score during colonoscopy. Standardized mean differences are shown with 95% CIs.

0.77; z = 2.82; P < 0.005). Two included trials contributed to the combined calculation of pain score. There was significant heterogeneity among the trials (s2 = 0.05, v2 = 3.25, d.f. = 1, P = 0.07; I2 = 69%). In the random effects model CI and AI were associated with a similar pain score 24 h after colonoscopy (SMD 0.02; 95% CI 0.41 to 0.36; z = 0.12; P = 0.20).

There was significant heterogeneity in the 15 trials which compared this variable (s2 = 0.07, v2 = 61.27, d.f. = 14, P < 0.00001; I2 = 77%). In the random effects model CI was associated with quicker caecal intubation compared with AI (SMD 0.18; 95% CI 0.34 to 0.03; z = 2.27; P = 0.02).

Caecal intubation failure rate

Discussion

Twelve trials contributed to the combined calculation of this variable. There was no heterogeneity (s2 = 0.0, v2 = 9.17, d.f. = 10, P = 0.52; I2 = 0%). In the random effects model the risk of failed caecal intubation was similar in both groups (OR 0.96; 95% CI 0.63– 1.45; z = 0.21; P = 0.83).

Based upon the findings of this largest ever study of 3607 patients undergoing colonoscopy with CI and AI, the former seems to have clinical advantages in terms of reduced procedural and postprocedural pain with a similar caecal intubation rate and faster time to visualize the caecum. These findings are consistent with

Time to visualize the caecum

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

119

CO2 for colonoscopy

M. S. Sajid et al.

(a) Study or subgroup Bretthauer 2002 Bretthauer 2005 Cleland 2013 Liu 2009 Sumanac 2002 Wong 2008 Yamano 2010 Total (95% CI)

Odds ratio CO2 insufflation Air insufflation Events Total Events Total Weight M-H, Random, 95%CI 60 16 7 17 3 22 10

121 43 108 174 49 44 66 605

71 35 29 136 23 7 23

119 42 97 176 51 49 54

15.0% 14.0% 14.3% 14.9% 13.3% 14.1% 14.4%

0.66 [0.40, 1.11] 0.12 [0.04, 0.33] 0.16 [0.07, 0.39] 0.03 [0.02, 0.06] 0.08 [0.02, 0.29] 6.00 [2.22, 16.22] 0.24 [0.10, 0.57]

588 100.0%

0.24 [0.07, 0.85]

Total events 135 324 Heterogeneity Tauz = 2.69; Chiz = 102.74, df = 6 (P < 0.0001); Iz = 94% Test for overall effect: Z = 2.21 (P = 0.03) (b) Study or Subgroup Bretthauer 2002 Chen 2013 Cleland 2013 Diez-Redondo 2012 Geyer 2011 Imai 2013 Uraoka 2009 Wong 2008 Yamano 2010

Air insufflation CO2 insufflation Mean SD Total Mean SD Total 1.7 0.5 119 0.2 0.5 121 97 1 96 1 0 0 97 0.17 0.85 108 0.73 1.37 0.5 129 2.18 0.5 141 0.81 2 2.3 109 0.6 1.2 110 1.3 0.5 18 19 0.5 0.5 57 57 1.8 2 2 0.8 44 0.39 0.15 49 0.16 0.1 66 54 0.7 0.5 0.5 0.2

Std. Mean Difference IV, Random, 95% CI 11.2% –2.99 [–3.36, –2.62] 11.3% 0.00 [–0.28, 0.28] 11.3% –0.50 [–0.77, –0.22] 11.3% –2.73 [–3.07, –2.40] 11.3% –0.76 [–1.04, –0.49] 10.2% –0.76 [–2.31, –0.82] 11.2% –0.50 [–0.87, –0.12] 11.0% –1.77 [–2.25, –1.29] 11.2% –0.99 [–1.38, –0.61]

Weight

Total(95% CI) 750 741 100.0% –1.31 [–2.02, –0.59] Heterogeneity: Tauz = 1.14; Chiz = 294.18, df = 8 (P < 0.00001); Iz = 97% Test for overall effect: Z = 3.59 (P = 0.0003)

Odds ratio M-H, Random, 95% CI

0.02 0.1 1 10 50 Favours CI Favours AI Std. Mean Difference IV, Random, 95% CI

–2 –1 0 1 2 Favours CI Favours AI

Figure 4 (a) Forest plot for pain 1 h after colonoscopy. Odds ratios are shown with 95% CIs (on plot CI, CO2 insufflation; AI, air insufflation). (b) Forest plot for pain score 1 h after colonoscopy. Standardized mean differences are shown with 95% CI.

previously reported systematic reviews containing fewer RCT. A review by Wang et al. [50] reported the combined analysis of 13 RCTs of CI during colonoscopy, four RCTs of endoscopic retrograde cholangiopancreatography, two RCTs of double-balloon enteroscopy, one RCT of oesophagogastroduodenoscopy and one RCT of flexible sigmoidoscopy, and concluded that when compared with AI, CI was associated with lower postprocedural pain and bowel distension. Both Wu et al. [51] and Dellon et al. [52] reported the combined analysis of nine RCTs in two separate studies and concluded that CI could decrease abdominal discomfort during and following colonoscopy without any additional adverse reaction, warranting its routine clinical use. The RCTs [29–49] included in the present review evaluated procedure-related pain either as a primary or secondary outcome according to the trial protocol; this variable is a major source of morbidity and influences the quality of life in patients undergoing colonoscopy. It was thoroughly investigated in the included RCTs.

120

Based on randomization, allocation concealment, power calculation, single or double blinding and reporting of ITT, the majority of the included trials were considered methodologically adequate. The combined outcome of these trials, in conjunction with the results of previously published studies [50–52], demonstrates sufficient evidence to support the conclusion that CI seems to have clinically proven advantages over AI. The safety and efficacy of CI for colonoscopy has been confirmed in several RCTs [30,39–43]. Seo et al. [43] measured the end tidal volume of CO2 (ETCO2) in patients undergoing colonoscopy. The mean ETCO2 values measured immediately after the procedure were significantly higher in the CI than in the AI group, but the values for ETCO2 were within the normal range [43]. The effect of CI on the adenoma detection rate (ADR) has not yet reported, but a prospective RCT is currently under way (NCT01782014) to compare the ADR between water insufflation, CI and AI during minimal sedation colonoscopy [53].

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

CO2 for colonoscopy

M. S. Sajid et al.

(a) CO2 insufflation Air insufflation Odds ratio Total Events Total Weight M-H, Random, 95%CI Study or subgroup Events Bretthauer 2002 Bretthauer 2005 Chen 2013 Liu 2009 Riss 2009 Seo 2013 Stevenson 1992 Sumanac 2002 Yamano 2010

10 16 0 3 21 6 1 4 10

121 43 96 174 157 46 27 49 66

14.9% 13.4% 12.4% 15.7% 9.8% 7.7% 12.5% 13.6%

0.17 [0.08, 0.35] 0.12 [0.04, 0.33] Not estimable 0.05 [0.02, 0.17] 0.44 [0.24, 0.80] 3.45 [0.66, 18.06] 0.04 [0.00, 0.35] 0.19 [0.06, 0.63] 0.89 [0.33, 2.38]

759100.0% 0.25 [0.11, 0.55]

779

Total (95% CI)

119 42 97 176 143 48 29 51 54

42 35 0 44 37 2 14 16 9

Odds ratio M-H, Random, 95% CI

71

199 Total events Heterogeneity Tauz = 0.96; Chiz = 32.55, df = 7 (P < 0.0001); Iz = 78% Test for overall effect: Z = 3.43 (P = 0.0006)

0.005 0.1

1

10

200

Favours CI Favours AI (b) CO2 insufflation Study or subgroup

Mean

Amato 2013 Bretthauer 2002 Diez-Redondo 2012 Geyer 2011 Imai 2013 Uraoka 2009 Yamano 2010

3 0.25 0.36 0.6 0.2 0.1 0.15

Air insufflation

Std. Mean Difference

SD Total Mean

SD Total Weight

1 0.5 0.5 1.2 0.1 2 0.5

1 0.5 0.5 1.2 0.1 2 0.5

115 121 129 110 19 57 66

4.6 0.8 0.78 0.7 0.2 0.5 0.2

113 119 141 109 18 57 54

Std. Mean Difference

IV, Random, 95% CI

IV, Random, 95% CI

14.7% –1.59 [–1.89, –1.30] 14.9% –1.10 [–1.37, –0.82] 15.0% –0.84 [–1.09, –0.59] –0.08 [–0.35, 0.18] 14.9% 0.00 [–0.64, 0.64] 12.0% –0.20 [–0.57, 0.17] 14.2% –0.10 [–0.46, 0.26] 14.3%

Total(95% CI) 617 611 100.0% –0.58 [–1.04, –0.12] Heterogeneity: Tauz = 0.35; Chiz = 86.93, df = 6 (P < 0.00001); Iz = 93% –2 Test for overall effect: Z = 2.47 (P = 0.01)

–1 Favours CI

0

1

2

Favours AI

Figure 5 (a) Forest plot for pain 6 h after colonoscopy. Odds ratios are shown with 95% CIs (on plot CI, CO2 insufflation; AI, air insufflation). (b) Forest plot for pain score 6 h after colonoscopy. Standardized mean differences are shown with 95% CI.

The cost of CI needs to be evaluated before routine application. Prices for CO2 regulators and insufflators vary from $5000 to $8000 [53,54]. A special water bottle for connecting tubing costs approximately $450 [53–55]. The ongoing cost of CO2 insufflation relates to the gas itself; two of the three available systems use disposable canisters. The authors could not find any studies comparing the cost effectiveness of CI with AI. The present review has several limitations. There was statistically significant heterogeneity among the included trials. There are many causes of this which could be attributed to variance in clinical practice and differing methodologies. The major methodological flaws in the included trials were the lack of a uniform and standardized pain-measuring tool. The endoscopists performing colonoscopy were of variable experience and differences in operator-dependent pain score were not reported adequately. Studies containing a small number of patients may not have had sufficient power. The use of

sedation, including dose and the approach to minimal sedation, can alter the pain score significantly and this could lead to sample contamination and biased outcome. The total volume of CO2 used to complete a colonoscopy is influenced by many factors including the procedure, length of the colon, ventilatory rate, pain during the procedure and associated cardiorespiratory comorbidity. Unfortunately this variable was either insufficiently reported or inadequately investigated among included studies. The variable volumes of CO2 used in individual patients in a trial and then among all included trials may sway the readings of all analysed outcomes; therefore this confounding factor should always be kept in mind.

Author contributions Study conception: Mr MS Sajid, MBBS, MBA, MSc, FRCS, Specialist Registrar Colorectal Surgery; Mr MK

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

121

CO2 for colonoscopy

M. S. Sajid et al.

Baig, MBBS, MD, FRCS, Consultant Colorectal Surgeon. Mr WFA Miles, MBBS, FRCS, Consultant Colorectal Surgeon; Mr P Sains, MBBS, MRCS, Consultant Colorectal Surgeon. Design: Mr MS Sajid, Mr MI Bhatti, MBBS, FRCS, Specialist Registrar Colorectal Surgery; Mr WFA Miles and Mr P Sains. Literature search: Mr MS Sajid, Mr MI Bhatti and Ms J Caswell, MBBS, MRCS, Surgical Research Fellow. PRISMA flow chart for study selection: Mr MS Sajid, Mr MI Bhatti, Ms J Caswell and Mr MK Baig. Data acquisition: Mr MS Sajid, Mr I Bhatti and Ms J Caswell. Data confirmation: Mr MK Baig, Mr WFA Miles, Mr P Sains and Ms J Caswell. Data analysis: Mr MS Sajid, Mr MK Baig and Mr MI Bhatti. Data interpretation: Mr MS Sajid, Mr MK Baig, Mr WFA Miles and Mr MI Bhatti. Drafting the article: Mr MS Sajid, Mr MK Baig, Mr J Caswell and Mr MI Bhatti. Proofreading of the article: Mr MK Baig, Mr WFA Miles and Mr P Sains. Final approval of the version to be published: Mr MS Sajid, Mr MI Bhatti, Ms J Caswell, Mr WFA Miles, Mr MK Baig and Mr P Sains.

Conflict of interest None to declare.

Funding None to declare.

References 1 Atkin W, Dadswell E, Wooldrage K et al.; SIGGAR investigators. Computed tomographic colonography versus colonoscopy for investigation of patients with symptoms suggestive of colorectal cancer (SIGGAR): a multicentre randomised trial. Lancet 2013; 381: 1194–202. 2 Carroll M, Seaman H, Halloran S. Tests and investigations for colorectal cancer screening. Clin Biochem 2014; 47: 921–39.
Carrascosa-Gil 3 Fern
andez-Calder
on M, Mu~ noz-Navas MA, J et al. Carbon dioxide vs. air insufflation in ileo-colonoscopy and in gastroscopy plus ileo-colonoscopy: a comparative study. Rev Esp Enferm Dig 2012; 104: 237–41. 4 Hussein AM, Bartram CI, Williams CB. Carbon dioxide insufflation for more comfortable colonoscopy. Gastrointest Endosc 1984; 30: 68–70. 5 Eberl S, Preckel B, Fockens P, Hollmann MW. Analgesia without sedatives during colonoscopies: worth considering? Tech Coloproctol 2012; 16: 271–6. 6 Eckardt VF, Kanzler G, Schmitt T, Eckardt AJ, Bernhard G. Complications and adverse effects of colonoscopy with selective sedation. Gastrointest Endosc 1999; 49: 560–5. 7 Aljebreen AM, Almadi MA, Leung FW. Sedated vs unsedated colonoscopy: a prospective study. World J Gastroenterol 2014; 20: 5113–8.

122

8 Saunders BP, Elsby B, Boswell AM, Atkin W, Williams CB. Intravenous antispasmodic and patient-controlled analgesia are of benefit for screening flexible sigmoidoscopy. Gastrointest Endosc 1995; 42: 123–7. 9 Saunders BP, Williams CB. Premedication with intravenous antispasmodic speeds colonoscope insertion. Gastrointest Endosc 1996; 43: 209–11. 10 Rondonotti E, Radaelli F, Paggi S et al. Hyoscine N-butylbromide for adenoma detection during colonoscopy: a randomized, double-blind, placebo-controlled study. Dig Liver Dis 2013; 45: 663–8. 11 Devitt J, Shellman L, Gardner K, Nichols LW. Using positioning after a colonoscopy for patient comfort management. Gastroenterol Nurs 2011; 34: 93–100. 12 Falt P, Liberda M, Smajstria V et al. Combination of water immersion and carbon dioxide insufflation for minimal sedation colonoscopy: a prospective, randomized, single center trial. Eur J Gatsroenterol Hepatol 2012; 24: 971–7. 13 Leung CW, Kaltenbach T, Soetikno R, Wu KK, Leung FW, Friedland S. Water immersion versus standard colonoscopy insertion technique: randomized trial shows promise for minimal sedation. Endoscopy 2010; 42: 557–63. 14 Pohl J, Messer I, Behrens A, Kaiser G, Mayer G, Ell C. Water infusion for cecal intubation increases patient tolerance, but does not improve intubation of unsedated colonoscopies. Clin Gastroenterol Hepatol 2011; 9: 1039–43. 15 Leung FW, Hu B, Wu J. Comparative effectiveness of water immersion and water exchange versus air insufflation for colonoscopy. J Interv Gastroenterol 2013; 3: 100–3. 16 Becker GL. The prevention or gas explosions in the large bowel during electrosurgery. Surg Gynecol Obstet 1953; 97: 463–7. 17 Williams CB. Who’s for CO2? Gastrointest Endosc 1986; 32: 365–7. 18 Geyer M, Guller U, Beglinger Ch. Carbon dioxide insufflation in colonoscopy is safe: a prospective trial of 347 patients. Diagn Ther Endosc 2012; 2012: 692532. 19 Higgins JPT, Green S (eds). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.0 (updated February 2008). http://www.cochrane-handbook.org (accessed May 2014). 20 Review Manager (RevMan) [Computer program]. Version 5.2.12. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2008. 21 DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986; 7: 177–88. 22 DeMets DL. Methods for combining randomized clinical trials: strengths and limitations. Stat Med 1987; 6: 341–50. 23 Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539–58. 24 Egger M, Smith GD, Altman DG. Systematic Reviews in Healthcare. London, BMJ Publishing, 2006. 25 Deeks JJ, Altman DG, Bradburn MJ. Statistical methods for examining heterogeneity and combining results from several studies in meta-analysis. In: Systemic Reviews in Health Care: Meta-Analysis in Context, 2nd edition. (eds

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

CO2 for colonoscopy

M. S. Sajid et al.

26

27

28 29

30

31

32

33

34

35

36

37

38

39

Egger M, Smith MD, Altman DG). London: BMJ Publication group, 2001. pp. 285–312. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 1–12. Chalmers TC, Smith H Jr, Blackburn B, Silverman B, Schroeder B, Reitman D. A method for assessing the quality of a randomized control trial. Control Clin Trials 1981; 2: 31–49. Cochrane IMS. http://ims.cochrane.org/revman/otherresources/gradepro/download (accessed May 2014). Amato A, Radaelli F, Paggi S, Baccarin A, Spinzi G, Terruzzi V. Carbon dioxide insufflation or warm-water infusion versus standard air insufflation for unsedated colonoscopy: a randomized controlled trial. Dis Colon Rectum 2013; 56: 511–8. Bretthauer M, Thiis-Evensen E, Huppertz-Hauss G et al. NORCCAP (Norwegian colorectal cancer prevention): a randomised trial to assess the safety and efficacy of carbon dioxide versus air insufflation in colonoscopy. Gut 2002; 50: 604–7. Bretthauer M, Hoff GS, Thiis-Evensen E, Huppertz-Hauss G, Skovlund E. Air and carbon dioxide volumes insufflated during colonoscopy. Gastrointest Endosc 2003; 58: 203–6. Bretthauer M, Lynge AB, Thiis-Evensen E, Hoff G, Fausa O, Aabakken L. Carbon dioxide insufflation in colonoscopy: safe and effective in sedated patients. Endoscopy 2005; 37: 706–9. Chen PJ, Li CH, Huang TY et al. Carbon dioxide insufflation does not reduce pain scores during colonoscope insertion in unsedated patients: a randomized, controlled trial. Gastrointest Endosc 2013; 77: 79–89. Church J, Delaney C. Randomized, controlled trial of carbon dioxide insufflation during colonoscopy. Dis Colon Rectum 2003; 46: 322–6. Cleland A, Carryer J, La Grow S. Carbon dioxide insufflation during colonoscopy: a randomised controlled trial. N Z Med J 2013; 126: 87–94. Diez-Redondo P, Gil-Simon P, Alcaide-Suarez N et al. Comparison between insufflation with air or carbon dioxide during the colonoscopy in sedated patients with propofol. Rev Esp Enferm Dig 2012; 104: 411–7. Geyer M, Guller U, Beglinger C. Carbon dioxide insufflation in routine colonoscopy is safe and more comfortable: results of a randomized controlled double-blinded trial. Diagn Ther Endosc 2011; 2011: 378906. Iida T, Okamura S, Kakizaki S et al. Carbon dioxide insufflation reduces the discomfort due to colonoscopy as objectively analyzed by salivary stress markers. Acta Gastroenterol Belg 2013; 76: 219–24. Imai A, Kato M, Ono S, Shimizu Y, Takeda H, Asaka M. Efficacy of carbon dioxide-insufflating colonoscopy in patients with irritable bowel syndrome: a randomized double-blind study. J Gastroenterol Hepatol 2012; 27: 1623–8.

40 Liu X, Liu D, Li J, Ou D, Zhou Z. Safety and efficacy of carbon dioxide insufflation during colonoscopy. Zhong Nan Da Xue Xue Bao Yi Xue Ban 2009; 34: 825–9. 41 Mayr M, Miller A, Gauger U, Rosch T. CO2 versus air insufflation for private practice routine colonoscopy: results of a randomized double blind trial. Z Gastroenterol 2012; 50: 445–8. 42 Riss S, Akan B, Mikola B et al. CO2 insufflation during colonoscopy decreases post-interventional pain in deeply sedated patients: a randomized controlled trial. Wien Klin Wochenschr 2009; 121: 464–8. 43 Seo EH, Kim TO, Park MJ et al. The efficacy and safety of carbon dioxide insufflation during colonoscopy with consecutive esophagogastroduodenoscopy in moderately sedated outpatients: a randomized, double-blind, controlled trial. J Clin Gastroenterol 2013; 47: e45–9. 44 Singh R, Neo EN, Nordeen N et al. Carbon dioxide insufflation during colonoscopy in deeply sedated patients. World J Gastroenterol 2012; 18: 3250–3. 45 Stevenson GW, Wilson JA, Wilkinson J, Norman G, Goodacre RL. Pain following colonoscopy: elimination with carbon dioxide. Gastrointest Endosc 1992; 38: 564–7. 46 Sumanac K, Zealley I, Fox BM et al. Minimizing postcolonoscopy abdominal pain by using CO(2) insufflation: a prospective, randomized, double blind, controlled trial evaluating a new commercially available CO(2) delivery system. Gastrointest Endosc 2002; 56: 190–4. 47 Uraoka T, Kato J, Kuriyama M et al. CO(2) insufflation for potentially difficult colonoscopies: efficacy when used by less experienced colonoscopists. World J Gastroenterol 2009; 15: 5186–92. 48 Wong JC, Yau KK, Cheung HY, Wong DC, Chung CC, Li MK. Towards painless colonoscopy: a randomized controlled trial on carbon dioxide-insufflating colonoscopy. ANZ J Surg 2008; 78: 871–4. 49 Yamano HO, Yoshikawa K, Kimura T et al. Carbon dioxide insufflation for colonoscopy: evaluation of gas volume, abdominal pain, examination time and transcutaneous partial CO2 pressure. J Gastroenterol 2010; 45: 1235–40. 50 Wang WL, Wu ZH, Sun Q et al. Meta-analysis: the use of carbon dioxide insufflation vs. room air insufflation for gastrointestinal endoscopy. Aliment Pharmacol Ther 2012; 35: 1145–54. 51 Wu J, Hu B. The role of carbon dioxide insufflation in colonoscopy: a systematic review and meta-analysis. Endoscopy 2012; 44: 128–36. 52 Dellon ES, Hawk JS, Grimm IS, Shaheen NJ. The use of carbon dioxide for insufflation during GI endoscopy: a systematic review. Gastrointest Endosc 2009; 69: 843–9. 53 http://clinicaltrials.gov/show/NCT01782014 (accessed May 15 2014). 54 http://www.asge.org/uploadedFiles/methodsofluminaldistensionforcolonoscopy (accessed July 2014). 55 Janssens F, Deviere J, Eisendrath P et al. Carbon dioxide for gut distension during digestive endoscopy: technique and practice survey. World J Gastroenterol 2009; 15: 1475–9.

Colorectal Disease ª 2014 The Association of Coloproctology of Great Britain and Ireland. 17, 111–123

123

Copyright of Colorectal Disease is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.