Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Dec 1.
Published in final edited form as: Gastrointest Endosc. 2017 Aug 9;86(6):972–985.e3. doi: 10.1016/j.gie.2017.07.046

Randomized sham-controlled trials in endoscopy: a systematic review and meta-analysis of adverse events

Allison R Schulman 1,2, Violeta Popov 3, Christopher C Thompson 1,2
PMCID: PMC5693737  NIHMSID: NIHMS898764  PMID: 28802556

Abstract

Background and Aims

Sham procedures in endoscopy are used with the intention of controlling for placebo response, potentially allowing more precise evaluation of treatment effect. Nevertheless, this type of study may impose significant risk without potential benefit for those in the sham group. The aim of the current study is to systematically review and analyze the endoscopic literature to assess the safety of sham controls.

Methods

MEDLINE and Embase databases were searched for endoscopic sham procedures for all dates through July 2017. Only randomized controlled trials comparing an endoscopic therapy with a sham were included. Primary outcome was adverse events (AE) categorized as mild, moderate, or severe. Results were combined using a random-effects model. Heterogeneity was assessed with I-squared, and publication bias was assessed with the Egger test and funnel plots.

Results

Data were extracted from 34 publications (1987–2017; 100% full-text), with a total of 2,492 procedures (1,355 treatment/1,137 sham). Sham procedures involved upper endoscopy (31 studies) and ERCP (3 studies). Treatment arms included procedures with the following indications: weight loss (38.2%), GI bleeding (26.5%), GERD (20.6%), sphincter of Oddi dysfunction (SOD) (8.8%), and dysphagia (6.2%). Overall percentage of severe adverse events (SAE) in the sham group was 1.7% (19/1137). Of these, the most common SAE in sham groups were need for surgery/intensive care unit (ICU) stay (35.3%), post-ERCP pancreatitis (23.5%), and perforation (11.8%). There was no significant difference in the odds of developing a severe adverse event between the treatment group and the sham group (OR, 1.3; 95% CI, 0.7 – 2.3]).

Conclusion

The frequency of adverse events in endoscopic sham procedures is substantial, and subjects patients to considerable morbidity. These results raise a serious ethical dilemma regarding the use of sham-controlled trials.

INTRODUCTION

Sham-controlled trials have become increasingly common in endoscopic research. Sham procedures are used with the intention of controlling for placebo response, potentially allowing more precise evaluation of treatment effect. Moreover, it may increase the scientific validity of an RCT1–3. Often, the use of sham-controlled trials, opposed to non-sham controls, is a requirement for device approval or procedure reimbursement.

Sham-controlled trials may be of additional benefit to the patients involved. Assignment to the control group may allow subjects involved in such research to receive standard medical treatment at no cost. These patients may be eligible to undergo the intervention at a later date should it prove to be beneficial, and, conversely, provides the hindsight to avoid the intervention if found to be harmful.

However, sham procedures pose an indisputable ethical dilemma4,5. Opponents of sham procedures argue that this type of study may impose significant risk without potential benefit for those in the sham group. Sham endoscopic interventions are not completely innocuous, and may result in more than minimal risks to participants. Furthermore, participation in this research may constitute an unacceptable violation of the rights of research subjects, with the potential for unforeseeable risks, and without benefits for those in the sham group.

The aim of the current study was to systematically review and analyze the endoscopic literature to assess the safety of sham controls.

METHODS

Literature Search

Before initiating the literature search, the search strategy, study inclusion and exclusion criteria, primary and secondary outcomes, and planned statistical analyses were defined and are described below.

MEDLINE and Embase databases were searched for endoscopic sham procedures from inception through July 2017 using the Ovid interface. Studies in English with greater than 5 adult participants were included. Animal trials were excluded. A search strategy using a combination of subject headings and text words was constructed to find articles that included endoscopic trials in which there was a sham control arm.

Review of titles/abstracts, full review of potentially relevant studies, and data abstraction were performed independently and in parallel by 2 authors (V.P. and A.S.). Only randomized controlled trials comparing an endoscopic therapy to a sham were included. The lists of full manuscripts meeting inclusion criteria from the 2 reviewers (V.P. and A.S.) were compared, and any disagreements were resolved by discussion and consensus, with a senior author (C.T.) serving as the final arbiter if consensus was not achieved.

Study Selection

Study Design and Population

We included all randomized controlled trials (RCTs) comparing an endoscopic intervention to an endoscopic sham control arm. Although no consensus definition exists for a sham procedure in endoscopy, an adopted definition from the Oxford English Dictionary was used in this analysis6. We defined an endoscopic sham procedure as one performed on a control group subject to expose him or her to similar incidental effects as those experienced by study group subjects on whom the investigational endoscopic intervention is performed. All observational studies and randomized controlled trials that did not include an endoscopic sham arm were excluded. Additionally, studies describing interventional therapy without providing data on adverse events were excluded.

Intervention

The study therapy was defined as either upper endoscopy, colonoscopy, or endoscopic retrograde cholangiopancreatography (ERCP) performed in both the treatment and sham arms, with a difference in intervention performed in the former group. These interventions included but were not limited to endoscopic bariatric therapies, Barrett’s ablation procedures, reflux procedures, treatment of gastrointestinal hemorrhage, and procedures to treat sphincter of Oddi dysfunction.

Comparison

Odds ratios (ORs) and risk ratios (RRs) in the outcome parameter after the endoscopic procedure were pooled together, weighted and compared. Additionally, 95% confidence intervals (CIs), event rate, heterogeneity, and variance between sham and treatment groups when comparing mild, moderate, and severe adverse events were evaluated.

Outcomes

The primary outcome was adverse events categorized as mild (heartburn, sore throat, transient epigastric pain, or bloating), moderate (dysphagia, vomiting, desaturation, stenosis, ulceration, severe reflux, post-ERCP pancreatitis, or hospitalization), or severe (bleeding requiring intervention, aspiration pneumonia, perforation, sepsis, organ failure, or other adverse event requiring surgery or any of the above adverse events requiring ICU stay). These predefined outcomes were chosen as they characterize the clinically relevant spectrum of adverse events incurred by undergoing endoscopic procedures. The results are reported as pooled adverse events and subgroup analyses based on categories. The secondary outcome was to determine the added risk incurred by subjects randomized to the sham arm, followed by crossover to treatment, in studies with a partial cross-over design.

Data Extraction

A data collection sheet was created to record information on study characteristics and outcomes. Data were extracted by 2 authors (A.S. and V. P.) independently and then compared. Any disagreements were resolved by discussion with the senior author (C.T.).

Data Analysis

For all of our studies, we calculated the pooled ORs and RRs for mild, moderate, and severe adverse events for both the sham and treatment groups, in addition to the 95% confidence interval, event rate, and P value. We pooled the results of the RCTs to assess the magnitude of the effect.

Risk-of-bias assessment of the RCTs was done using the Cochrane Collaboration Risk-of-Bias tool7. We extracted data regarding quality based on several domains: sequence generation, allocation concealment, blinding of participants and outcome assessment, incomplete data, and selective reporting.

Results were combined using a random-effects analysis due to our initial assumption of variation among the individual studies. Heterogeneity among the individual studies was assessed by the Cochrane Q-test and the I2 statistic. Heterogeneity was defined as significant if I2 was greater than 50%. A funnel plot was constructed to evaluate for publication bias for the main outcomes. The funnel plot was inspected visually for asymmetry. Egger testing was also performed. In this study, acuity of illness and year of study were thought to be possible sources of heterogeneity. If substantial heterogeneity was found for the primary outcomes, sensitivity analysis was performed by acuity of illness as evidenced by indication and type of the intervention. Patients undergoing urgent procedures for upper GI/variceal bleeding were considered to be of poorer health status than other populations undergoing non-urgent procedures. Additionally, studies were analyzed by year (≤2005 vs >2005). All analyses were done using Comprehensive Meta-Analysis software, version 2.0 (Biostat, Inc Englewood, NJ, USA) unless specified otherwise.

RESULTS

Study Selection

We identified 889 citations that matched our initial search criteria, and another 7 citations from other sources. After removal of duplicates, 416 articles remained and were reviewed (Figure 1). After abstract review, 53 studies were identified. After full manuscript review and discussion, thirteen additional studies were excluded given the intervention was not endoscopic in nature. Authors with insufficient information regarding adverse events were contacted, and those in which information remained unavailable were also excluded8–12. Two final studies were excluded on the basis of including fewer than three participants in the sham arm. After review, 34 publications between 1987 and 2017 were included from which data were extracted. All of these publications were full-text.

Figure 1.

Figure 1

Open in a new tab

PRISMA flow diagram of literature search results.

Study characteristics

Characteristics and outcomes of individual trials are described in Table 1. All 34 studies were randomized controlled trials. Trials were conducted in several countries, including the United States, Spain, Italy, Netherlands, Australia, Belgium, Germany, Sweden, and Singapore. Trials ranged from one to 15 centers, enrolling between 18 and 332 patients per trial. A total of 2,492 endoscopic procedures were performed, 1,355 from the treatment arms and 1,137 from the sham arms. Sham procedures involved upper endoscopy and ERCP in 31 and 3 studies, respectively, and consisted of sedation or general anesthesia, a minimum of one endoscopic procedure, and additional interventions including but not limited to mucosal fluid injection, installation of large fluid boluses into the stomach, overtube placement, and Bougie advancement with frequent rotation (Table 1).

Table 1.

Characteristics of studies included

Study Study Procedure Treatment intervention performed Sham procedure Origin of Study N Treatment N Sham
Fleischer, 198536 Endoscopy ND:YAG laser therapy for varices General or sedation USA 10 10
Laine, 198737 Endoscopy Multipolar electrocoagulation for bleeding ulcer General or sedation; heater probe held in gut lumen USA 21 23
Geenen, 198938 ERCP Sphincterotomy Biliary/pancreatic manometry; Sphincterotome held in gut lumen USA 23 24
Laine, 198939 Endoscopy Heater probe for bleeding ulcer General or sedation; heater probe held in gut lumen USA 38 37
Rutgeerts, 198940 Endoscopy Epinephrine injection General anesthesia or sedation; Belgium 25 25
Mathus-Vliegen, 199041 Endoscopy Intragastric balloon placement General or sedation; Sham balloon Netherlands 14 13
Fullarton, 199142 Endoscopy Heater probe for bleeding ulcer General or sedation; heater probe held in gut lumen UK 20 23
Veteran’s Affairs, 199135 Endoscopy Sclerotherapy of varices ~10 EGDs with placebo mucosal injection USA 143 138
Rutgeerts, 199343 Endoscopy Alcohol injection of bleeding ulcer General anesthesia or sedation Belgium 25 25
Veteran’s Affairs, 199444 Endoscopy Sclerotherapy ≥1 EGD; Placebo mucosal injection USA 122 131
Hartigan, 199734 Endoscopy Sclerotherapy ~10 EGDs with placebo mucosal injection USA 44 43
Scolapio, 200145 Endoscopy Balloon dilation Sedation, balloon catheter placed but not inflated USA 43 40
Corley, 200317 Endoscopy Radiofrequency ablation Stretta balloon inflation (repeated without injection) USA 31 25
Mathus-Vliegen, 200318 Endoscopy Intragastric balloon placement Sham balloon, 3 EGDs Netherlands 11 17
Gorelick, 200446 ERCP Botox injection Sclerotherapy needle injection into bowel lumen USA 12 14
Mathus-Vliegen, 200547 Endoscopy Intragastric balloon placement Sham balloon, 4 EGDs Netherlands 12 21
Deviere, 200516 Endoscopy Enteryx General anesthesia or sedation; Injection of co-polymer Europe 31 30
Genco, 200648 Endoscopy Intragastric balloon placement 2 EGDs Italy 16 16
Schwartz, 200749 Endoscopy EndoCinch Sedation; overtube in place for 40 mins Netherlands 18 15
Rothstein, 200633 Endoscopy Plication of cardia Plicator (45 Fr) placed over a wire and retroflexed USA/Europe 72 72
Tarnoff, 200950 Endoscopy Duodenojejunal bypass liner General anesthesia or sedation USA 20 4
Bajbouj, 200914 Endoscopy Argon plasma coagulation (APC) General anesthesia or sedation; simulated APC USA 11 10
Aziz, 201051 Endoscopy Stretta Sedation USA 24 12
Rodriques, 201052 Endoscopy Duodenojejunal bypass liner General anesthesia Chile 12 3
Fockens, 201013 Endoscopy Gatekeeper prosthesis General anesthesia or sedation; overtube placed USA/Europe 75 43
Gersin, 201153 Endoscopy Duodenojejunal bypass liner Sedation USA 21 24
Arts, 201254 Endoscopy Stretta General anesthesia or sedation USA 9 11
Mathus-Vliegen, 201355 Endoscopy Intragastric balloon placement Sham balloon placed, multiple blood draws; 2 EGDs Netherlands 16 21
Thompson, 201356 Endoscopy Transoral gastrojejunal outlet reduction General anesthesia; overtube placed for 30 min USA 48 27
Mathus-Vliegen, 201457 Endoscopy Intragastric balloon placement Sham balloon placed, multiple blood draws Netherlands 16 21
Eid, 201458 Endoscopy Stomaphyx General anesthesia; scope rotated for 30–40 min USA 55 31
Cotton, 201459 ERCP Sphincterotomy Biliary/pancreatic manometry; pancreatic duct stent USA 118 55
Hakansson, 201560 Endoscopy Transoral incisionless fundoplication General anesthesia; scope manipulation for 30 min Sweden 22 22
Sullivan, 201723 Endoscopy POSE™ General anesthesia; 54 Fr Bougie placed and rotated every 15 mins × 3 times USA 221 111
Open in a new tab

Treatment arms within the selected randomized controlled trials (RCTs) included procedures with the following indications: weight loss (38.2%), GI bleeding (26.5%), GERD (20.6%), SOD (8.8%), and dysphagia (6.2%). Average age in the treatment and sham groups were 48.1 and 48.3 years, respectively. Furthermore, 50.6% and 50.3% of patients were female in the treatment and sham groups, respectively.

Primary outcome

Adverse events by category (mild, moderate severe) for both the sham and treatment groups are shown in Table 2. Total number of all adverse events was 1,257 (50.4%). There were no deaths attributed to the study or sham procedure. The pooled ORs and RRs, 95% confidence intervals, heterogeneity, and variance between sham and treatment groups when comparing mild, moderate, and severe adverse events are shown in Table 3. The OR (95% confidence interval [CI]) of developing an adverse event categorized as mild, moderate, or severe were 2.7 (95% CI, 1.2–5.6), 3.8 (95% CI,1.6–9.1), and 1.3 (95% CI, 0.7–2.4), respectively.

Table 2.

Adverse events by study

Number of Adverse events in Treatment Group Number of Adverse events in Sham Group
Study Study Procedure Mild Moderate Severe Mild Moderate Severe
Fleischer, 198536 Endoscopy 0 0 3 0 0 2
Laine, 198737 Endoscopy 0 0 0 0 0 0
Geenen, 198938 ERCP 0 1 1 0 0 2
Laine, 198939 Endoscopy 0 7 1 0 0 0
Rutgeerts, 198940 Endoscopy 0 0 2 0 0 4
Mathus-Vliegen, 199041 Endoscopy 1 1 0 5 0 0
Fullarton, 199142 Endoscopy 0 0 1 0 0 0
Veteran’s Affairs, 199135 Endoscopy 107 106 0 25 4 1
Rutgeerts, 199343 Endoscopy 0 0 8 0 0 4
Veteran’s Affairs, 199444 Endoscopy 0 0 0 0 0 0
Hartigan, 199734 Endoscopy 0 0 0 0 0 0
Scolapio, 200145 Endoscopy 0 0 0 0 0 0
Corley, 200317 Endoscopy 4 5 0 1 0 1
Mathus-Vliegen, 200318 Endoscopy 5 2 0 3 1 0
Gorelick, 200446 ERCP 0 2 1 0 6 0
Mathus-Vliegen, 200547 Endoscopy 7 3 0 5 0 0
Deviere, 200516 Endoscopy 16 5 0 6 0 0
Genco, 200648 Endoscopy 13 0 0 4 0 0
Schwartz, 200749 Endoscopy 11 0 0 2 1 0
Rothstein, 200633 Endoscopy 39 6 2 12 3 0
Tarnoff, 200950 Endoscopy 16 5 0 0 0 0
Bajbouj, 200914 Endoscopy 7 0 0 6 0 0
Aziz, 201051 Endoscopy 12 3 0 4 0 0
Rodriques, 201052 Endoscopy 12 6 0 0 0 0
Fockens, 201013 Endoscopy 23 2 4 9 3 0
Gersin, 201153 Endoscopy 14 7 3 0 0 0
Arts, 201254 Endoscopy 3 0 0 2 0 0
Mathus-Vliegen, 201355 Endoscopy 2 2 0 0 0 0
Thompson, 201356 Endoscopy 19 8 1 14 6 0
Mathus-Vliegen, 201457 Endoscopy 1 2 0 0 0 0
Eid, 201458 Endoscopy 62 16 2 29 4 2
Cotton, 201459 ERCP 0 15 2 0 10 2
Hakansson, 201560 Endoscopy 21 21 0 20 15 0
Sullivan, 201723 Endoscopy 208 74 2 93 10 1
Open in a new tab

Table 3.

Pooled odds ratio and risk ratio, 95% confidence intervals, heterogeneity, and variance between sham and treatment groups when comparing mild, moderate, and severe adverse events.

Adverse events Event Rate 95% Confidence Interval (CI) I-squared Tau-squared
Mild
 Sham 0.29 0.18–0.44 88.3 1.87
 Treatment 0.24 0.14–0.37 89.7 2.64
 Odds Ratio (OR) OR = 2.7 1.4–5.4 75.3 1.45
 Risk Ratio (RR) RR = 1.5 1.2–1.9 79.0 0.10
Moderate
 Sham 0.05 0.03–0.09 64.4 1.25
 Treatment 0.14 0.09–0.21 86.3 1.50
 Odds Ratio (OR) OR = 4.0 1.9–8.4 70.3 1.96
 Risk Ratio (RR) RR = 2.67 1.6–4.7 64.5 0.72
Severe
 Sham 0.04 0.03–0.06 14.7 0.21
 Treatment 0.04 0.03–0.07 48.1 0.80
 Odds Ratio (OR) OR = 1.3 0.7–2.3 0 0
 Risk Ratio (RR) RR = 1.3 0.7–2.1 0 0
Open in a new tab

The forest plot showing event rate of all adverse events for sham procedures is shown in Figure 2A. Total number of mild, moderate, and severe adverse events in sham groups was 240, 63, and 19, respectively. The overall percentage of SAE in the sham group was 1.7% (19/1137). Of these, the most common SAEs were needed for surgery/ICU stay (35.3%), post-ERCP pancreatitis (23.5%), and perforation (11.8%). The forest plot showing event rate of all adverse events for treatment procedures is shown in Figure 2B. Total number of mild, moderate, and severe adverse events in the treatment group was 603, 299, and 33, respectively. The overall percentage of SAE in the treatment group was 2.4% (33/1112). Of these, the most common SAEs were need for surgery/ICU stay (38.7%), perforation (16.1%), and bleeding (9.7%) were most common. Event rate, pooled odds ratio, 95% confidence intervals, heterogeneity, and variance between these groups are shown in Table 1. There was no significant difference in the odds of developing a severe adverse event between the treatment group and the sham group (OR 1.3 [0.7, 2.3]). The pooled OR [95% confidence interval] of developing an adverse event was 0.5 [0.3, 0.8] in the sham group compared with the treatment group. The forest plot comparing pooled ORs of adverse events for sham and treatment groups is shown in Figure 3.

Figure 2.

Figure 2

Figure 2

Open in a new tab

Forest plot showing mild, moderate, and severe adverse events for (A) sham and (B) treatment groups. *3 studies in (A)35,58,60 and 7 studies in (B)23,35,50,52,53,58,60 were excluded from these figures because the number of events was higher than the total number of subjects or there were zero adverse events reported in both sham and treatment AE categories.

Figure 3.

Figure 3

Open in a new tab

Forest plot showing pooled odds ratio of adverse events comparing sham and treatment groups. *10 studies in (B)23,34,35,37,45,50,52,53,58,60 were excluded from these figures because the number of events was higher than the total number of subjects or there were zero adverse events reported in both sham and treatment AE categories.

Sensitivity analyses were performed to investigate heterogeneity in mild and moderate pooled event rates for both sham and treatment arms. Acuity of intervention (ie, urgent versus non-urgent) had a significant impact on between study variability for sham pooled event rates for mild (R-squared =0.37) and severe (R-squared = 0.86) but could not explain the heterogeneity found in the treatment group. Additionally, a significantly increased event rate was found in studies performed >2005 versus ≤2005 (pooled mean = 0.16 vs 0.06, P value <0.001). The heterogeneity in pooled moderate AE could be explained by acuity and year of study, with a combined R-squared of 0.70.

Secondary outcome

Six studies reported adverse event data for subjects randomized to the sham arm, followed by cross-over to treatment13–18. In partial cross-over design studies, the estimated average pooled increased risk of developing a non-severe event after being randomized to the sham arm was risk ratio = 1.33 (95% CI, 1.14 – 1.56), p<0.001, I-squared = 6.74, Tau-squared = 0.003, compared with those randomized to the treatment group.

Quality Assessment and Publication Bias

A funnel plot was created to evaluate for publication bias. Visual inspection of the funnel plots revealed no obvious asymmetry to suggest publication bias, and there was no evidence of publication bias by the Egger test (Egger regression intercept = −1.3, p=0.23) (Figure 4). Biases for individual studies and overall quality are shown in Table 4.

Figure 4.

Figure 4

Open in a new tab

Funnel plot demonstrating low chance of publication bias.

Table 4.

Biases for individual studies and overall assessment of quality of manuscripts included.

Random sequence generation (selection bias) Allocation concealment (selection bias) Blinding of participants and personnel (performance bias) Blinding of outcome assessment (detection bias) Incomplete outcome data (attrition bias) Selective reporting (reporting bias) Overall quality
Fleischer, 198536 1985 + + + + − − Mod
Laine, 198737 1987 + + + + + + High
Geenen, 198938 1989 + + + + + + High
Laine, 198939 1989 + + + + + + High
Rutgeerts, 198940 1989
Mathus-Vliegen, 199041 1990 + + + + − + High
Fullarton, 199142 1991 − + + + + − Mod
Veteran’s Affairs, 199135 1991 + + + + + + High
Rutgeerts, 199343 1993 + + + + + + High
Veteran’s Affairs, 199444 1994 + + + + + + High
Hartigan, 199734 1997 − + + + + + High
Scolapio, 200145 2001 − + + − + + Mod
Corley, 200317 2003 + + + + + + High
Mathus-Vliegen, 200318 2003 − + + + − + Mod
Gorelick, 200446 2004 − + + + + + High
Mathus-Vliegen, 200547 2005 + + + + − + High
Deviere, 200516 2005 + + + + + + High
Genco, 200648 2006 + + + + + + High
Schwartz, 200749 2007 + + + + + + High
Rothstein, 200633 2006 + + + + + + High
Tarnoff, 200950 2009 + + + + + + High
Bajbouj, 200914 2009 + − + + + + High
Aziz, 201051 2010 + + + + + + High
Rodriques, 201052 2010 + + + + + + High
Fockens, 201013 2010 − + + + + + High
Gersin, 201153 2011 + + + + + + High
Arts, 201254 2012 − + + + + + High
Mathus-Vliegen, 201355 2013 − + + + + + High
Thompson, 201356 2013 + + + + + + High
Mathus-Vliegen, 201457 2014 − + + + + + High
Eid, 201458 2014 + + + + + + High
Cotton, 201459 2014 + + + + + + High
Hakansson, 201560 2015 + + + + + + High
Sullivan, 201723 2017 + + + + + + High
Open in a new tab

DISCUSSION

Sham-controlled trials have become increasingly more common in endoscopic research. This is in part due to the fact that endoscopic procedures are often amenable to sham designs, as there are no obvious incisions incurred during the procedure making it relatively easy to maintain blinding. Nevertheless, the use of sham procedures in human research raises ethical concerns, given the potential to harm subjects despite not receiving the desired intervention. Additionally, some endoscopic sham procedures are intended to induce symptoms that are suggestive of the patient having had the procedure, potentially carrying increased risk.

Many endoscopic trials often involve partial cross-over designs, which allow patients in the sham arm to receive the study procedure after unblinding. This is intended to have several positive effects including giving a sense of fairness to patients, increasing enrollment, and keeping patients who feel they have received the sham procedure engaged throughout the trial. Unlike with double cross-over pharmaceutical trials, the treatment group does not undergo a sham procedure following the unblinding process in the majority of these trials. This disparity creates potential for increased risk in those randomized to the sham arm, as many will typically undergo an additional procedure.

To our knowledge, this is the only meta-analysis to date investigating the adverse effects endured by the sham arm during endoscopic randomized controlled studies (RCTs). In the 34 trials that met inclusion criteria for this meta-analysis, there were a total number of 321 adverse events in the sham arm (240 mild, 63 moderate, and 18 severe). These numbers do not reflect adverse events incurred in the cross-over portion of the study where patients would receive the therapy under investigation. The most common severe adverse events in sham group were not insignificant, including six patients who required surgery/ICU stay, 4 patients who developed post-ERCP pancreatitis, and 2 patients who developed a perforation from the sham procedure itself, despite not undergoing the intervention planned for the treatment arm. Although pooled odds ratio of adverse events was 2 times greater in the treatment group than the control group, the risk of having an adverse event in the sham group was not trivial. In fact, there was no significant difference in SAE between groups (OR 1.3; 95% CI, 0.7 – 2.3). Thus, the risk of a sham procedure outweighs the absolute lack of clinical benefit, and may, in fact, exceed acceptable research risk.

Furthermore, subjects are not merely randomized to the study intervention or sham arm. Many subjects after sham procedures subsequently undergo a study intervention, as part of a partial cross-over design. This is intended to retain patients in the control arm and limit loss to follow-up. Unfortunately, numerous studies do not report the results of this partial cross-over. In our analysis, only six of 34 studies report cross-over outcomes. These subjects are not only exposed to the risk of the sham procedure, but also to the risk of the study intervention. The pooled additional risk incurred from being initially randomized to the sham arm and then receiving a cross-over intervention was significant (RR=1.33; 95% CI, 1.14, 1.56; p<0.001), compared with patients initially randomized to the study intervention. Thus, in partial cross-over designs, patients initially randomized to the sham group again incur disproportionately increased risk.

Additionally, sensitivity analysis was performed to clarify heterogeneity where present. We found that procedural acuity had an effect on heterogeneity in the sham group. This finding brings into question the ethics of withholding all treatment (for the purposes of research) in patients requiring urgent care. Furthermore, studies performed after 2005 contained more adverse events than older studies. This observation could be multi-factorial in origin. The nature of the sham procedure itself may be becoming more aggressive, as seen in obesity and GERD studies that require Bougie placement and manipulation (Table 1). Additionally, more meticulous reporting of adverse events may partially explain this finding. Regardless, the safety of endoscopic sham procedures does not appear to have improved over the last 30 years.

The results of this meta-analysis raise a serious ethical dilemma regarding the use of sham-controlled trials, opposed to non-sham randomized controls, as a requirement for endoscopic device approval or procedure reimbursement. Many have advocated that the risk of sham interventions may not be acceptable or reasonable in relation to the added knowledge provided due to the sham procedure itself4,19. Moreover, inclusion of a sham arm may lead to the risk of exploitation of participants for the sake of scientific knowledge, and may not be relevant to real world clinical practice20.

There are many different types of RCT design that involve random allocation to different groups of therapy. Subjects can be randomized to criterion standard, alternative therapies, no therapy with observation, or placebo/sham. Originally, placebo-controlled trials were designed for medication studies and introduced no increased risk to subjects beyond lack of an alternative therapy. Sham trials are similar to placebo controls except that they introduce a procedure in place of a sugar pill, therefore subjecting the sham arm to the risk associated with the sham procedure. Justification for this is predicated on the risk of a sham procedure not exceeding the threshold of acceptable research risk. Objections to this type of trial by ethicists and proceduralists are due to a variety of reasons21,22. First, there is substantial risk in an RCT that the intervention arm may not be effective, and could carry unknown morbidity. Having the control arm be a sham procedure additionally removes the possibility of being allocated to an arm with known benefit. This exposes trial patients to prolonged periods of time in which no subject may receive effective therapy, while incurring risk of the study procedure and/or the sham procedure. Second, the risk posed to the sham arm by the placebo intervention itself, knowing there is no clinical benefit, is of dubious ethical justification. Third, the use of active misleading aimed at making subjects believe that they are receiving the actual intervention, when they are in fact not, is of ethical concern21,22. Finally, some sham controls with partial cross-over design decrease scientific validity by introducing perverse incentives. For example, in the ESSENTIAL trial, subjects were randomized to an endoscopic obesity therapy or a sham procedure, with a partial cross-over for patients initially allocated to the sham arm23. The only subjects eligible for the cross-over study were those who did not lose weight. Subjects that lost weight were not considered eligible for cross-over. As a result, patients that were close to this cut-off reported deliberate increase in food consumption to avoid exclusion from the cross-over procedure. This can artificially decrease the effectiveness of an intervention. It may also diminish scientific validity by introducing systematic bias against the study therapy. Similar to this, an inverse placebo effect can be seen in subjects that do not believe they received an intervention, thereby minimizing the difference between the sham and treatment arm.

There are, however, some potential advantages to sham-controlled trials. A comparison of the experimental intervention with a sham arm facilitates blinding of the patient, investigator, and other study personnel, thereby providing the ability to discern between specific effects of the intervention and other potential biases3,24–29. Additionally, inclusion of a sham group has been postulated to provide knowledge that will prevent the introduction of insufficiently proven, potentially risky, interventions into clinical practice and thereby prevent unnecessary, and often expensive, costs to the health care system27,30.

In the present review, patients randomized to undergo a sham procedure still endured the risk of undergoing sedation and an endoscopy or ERCP, often in conjunction with a second procedure that carried additional hazard. The risk of severe adverse events inherent to upper endoscopic examination under sedation is approximately one per thousand procedures31, with a mortality rate estimated to be as high as three per ten thousand cases32. These events are mostly related to cardiovascular events, although perforation also occurs. This does not include the risk of additional sham manipulations, which may involve dilation, ablation, and other interventions. Particularly, ERCP carries additional risk from adverse events specific to pancreatobiliary instrumentation, including pancreatitis, sepsis, and hemorrhage or retroperitoneal duodenal perforation after therapeutic procedures. In our analysis, and consistent with previous literature, post-ERCP pancreatitis was the most common etiology leading to a severe adverse event observed in patients undergoing ERCP. It is well known that this adverse event can be associated with significant morbidity and cost to the healthcare system. Studies investigating novel obesity and GERD treatments report placement of large Bougies up to 54F being advanced into the esophagus and rotated at regular intervals to simulate the study intervention, device placement and maneuvering23,33. This carries the risk of mucosal abrasion, bleeding, aspiration, and perforation. Other study designs require numerous sham procedures, as many as ten per subject in some trials, where a placebo was injected into esophageal mucosa as a sham in a variceal bleeding study34,35. This exposes the subject to numerous sedation events, in addition to withholding actual treatment for as long as 24 months.

Based on the results of this study, investigators should reconsider use of sham-controlled trial designs for endoscopic interventions, while maintaining an emphasis on sound RCT designs. If the decision is made to proceed with a sham arm, it must be clear exactly what clinically relevant information will be provided by the study design, above that which could be provided by a standard RCT design. Additionally, investigators must take every precaution to minimize the risks associated with the sham procedure, and ensure that the patients are fully aware of the additional risks imposed by potentially having both a sham procedure and subsequently the study treatment.

There are several strengths of this meta-analysis. To our knowledge, this is the first study to systematically review and analyze the endoscopic literature to assess the safety of sham controls. Thirty-four studies with over two thousand endoscopic procedures were included, thus limiting publication bias. Sound methodological design was employed, and only studies which provided sufficient information were included, with authors contacted individually for missing data. Additionally, all included trials were RCTs, which limited conceptual bias due to poor study design known to influence other types of studies. Finally, adverse events were analyzed by severity to provide a clinical context to the findings.

There are also limitations to our study. Reporting of adverse events varied, with some studies only reporting SAE. As a result, mild and moderate adverse events may have been underreported, suggesting there may have been additional adverse events in both groups that were not captured, and sham procedures may carry more risk than we identified in this meta-analysis. Nevertheless, we suspect that underreporting would have been similar between treatment and sham groups, and therefore unlikely to affect the pooled odds ratios. Additionally, identification or determination of adverse events can be subjective and introduce bias.

In conclusion, the results of this systematic review raise a considerable ethical dilemma regarding the use of sham-controlled trials, opposed to non-sham controls, as a requirement for endoscopic device approval or procedure reimbursement. The risk imposed on the sham group is not insignificant, and may be disproportionate to the treatment group when considering the commonly used partial cross-over design. Standard RCT designs with an appropriate control group should be preferred to this higher risk sham-control strategy.

Database: Ovid MEDLINE(R) <1946 to June Week 1 2017>, Embase <1974 to 2017 June 13>
Search Strategy:

  1. endoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (277809)

  2. limit 1 to english language (221870)

  3. limit 2 to human (201266)

  4. sham.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (163275)

  5. limit 4 to english language (153439)

  6. limit 5 to human (31896)

  7. 3 and 6 (207)

  8. endoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (277809)

  9. limit 8 to human (251214)

  10. colonoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (97346)

  11. limit 10 to english language (84927)

  12. limit 11 to human (79304)

  13. 6 and 12 (41)

  14. endoscopic.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (363400)

  15. limit 14 to english language (300775)

  16. limit 15 to human (275071)

  17. 6 and 16(309)

  18. ercp.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (24174)

  19. limit 18 to english language (20517)

  20. limit 19 to human (18775)

  21. 6 and 20 (32)

  22. limit 21 to “all adult (19 plus years)” [Limit not valid in Embase; records were retained] (30)

  23. limit 22 to humans (30)

  24. limit 23 to (clinical trial or randomized controlled trial) (14)

  25. 7 or 13 or 17 or 21 or 24 (427)

  26. placebo.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (576716)

  27. limit 26 to english language (538149)

  28. limit 27 to human (483799)

  29. 3 and 12 and 16 and 20 and 28 (3)

  30. 3 and 28 (3772)

  31. 20 and 28 (450)

  32. 12 and 28 (1103)

  33. 16 and 28 (4176)

  34. 30 or 31 or 32 or 33 (7117)

  35. remove duplicates from 25 (295)

  36. remove duplicates from 25 (295)

  37. endoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (277809)

  38. limit 37 to english language (221870)

  39. limit 38 to human (201266)

  40. sham.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (163275)

  41. limit 40 to english language (153439)

  42. limit 41 to human (31896)

  43. 39 and 42 (207)

  44. endoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (277809)

  45. limit 44 to human (251214)

  46. colonoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (97346)

  47. limit 46 to english language (84927)

  48. limit 47 to human (79304)

  49. 42 and 48 (41)

  50. endoscopic.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (363400)

  51. limit 50 to english language (300775)

  52. limit 51 to human (275071)

  53. 42 and 52 (309)

  54. ercp.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (24174)

  55. limit 54 to english language (20517)

  56. limit 55 to human (18775)

  57. 42 and 56 (32)

  58. limit 57 to “all adult (19 plus years)” [Limit not valid in Embase; records were retained] (30)

  59. limit 58 to humans (30)

  60. limit 59 to (clinical trial or randomized controlled trial) (14)

  61. 43 or 49 or 53 or 57 or 60 (427)

  62. placebo.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, sy, tn, dm, mf, dv, kw, fx] (576716)

  63. limit 62 to english language (538149)

  64. limit 63 to human (483799)

  65. 39 and 48 and 52 and 56 and 64 (3)

  66. 39 and 64 (3772)

  67. 56 and 64 (450)

  68. 48 and 64 (1103)

  69. 52 and 64 (4176)

  70. 66 or 67 or 68 or 69 (7117)

  71. remove duplicates from 61 (295)

  72. remove duplicates from 61 (295)

  73. egd.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (7400)

  74. limit 73 to english language (7196)

  75. limit 74 to human (6184)

  76. 6 and 75 (8)

  77. enteroscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (7342)

  78. 6 and 77(1)

  79. 6 and 77(1)

  80. esophagoscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (27352)

  81. limit 80 to english language (20167)

  82. limit 81 to human (18882)

  83. 6 and 82 (30)

  84. gastroscopy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (40355)

  85. 6 and 84 (27)

  86. eus.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (21586)

  87. 6 and 86 (20)

  88. intragastric balloon.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (1131)

  89. 6 and 88 (43)

  90. BARRX.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (137)

  91. 6 and 90 (2)

  92. 61 or 76 or 79 or 83 or 85 or 87 or 89 or 91 (469)

  93. 36 or 92 (469)

  94. manometry.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (47559)

  95. 6 and 94 (143)

  96. 93 or 95 (551)

  97. mucosal ablation.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (260)

  98. 6 and 97(0)

  99. radiofrequency ablation.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (41330)

  100. 6 and 99 (101)

  101. 96 or 100 (621)

  102. esd.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (9406)

  103. 6 and 102 (3)

  104. emr.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (14583)

  105. 6 and 104(11)

  106. radiofrequency ablation.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (41330)

  107. 6 and 106 (101)

  108. cryotherapy.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (26660)

  109. 6 and 108 (46)

  110. argon plasma coagulation.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx](5116)

  111. 6 and 110(14)

  112. stent.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (208945)

  113. 6 and 112 (79)

  114. duodenal jejunal bypass liner.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (147)

  115. 6 and 114 (2)

  116. endobarrier.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (189)

  117. 6and116 (11)

  118. duodenal.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (141347)

  119. 6 and 118 (268)

  120. gastric.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (537887)

  121. 6 and 120 (886)

  122. esophageal.mp. [mp=ti, ab, ot, nm, hw, kf, px, rx, ui, an, sy, tn, dm, mf, dv, kw, fx] (240209)

  123. 6 and 122 (223)

  124. 101 or 103 or 105 or 107 or 109 or 111 or 113 or 115 or 117 or 119 or121 or123 (1628)

  125. 101 or 103 or 105 or 107 or 109 or 111 or 113 or 115 or 117 or 119 or 121 or 123 (1628)

  126. remove duplicates from 125 (1168)

  127. 96 or 126 (1341)

  128. remove duplicates from 127 (1168)
Open in a new tab

Abbreviations

RCT

Randomized controlled trial

AE

Adverse events

ERCP

Endoscopic retrograde cholangiopancreatography

OR

Odds ratio

ICU

Intensive care unit

GI

gastrointestinal

GERD

gastroesophageal reflux disease

CI

Confidence interval

Footnotes

Author Contributions:

Schulman AR: study concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; statistical analysis

Popov V: study concept and design; analysis and interpretation of data; critical revision of the manuscript for important intellectual content; study supervision

Thompson CC: study concept and design; critical revision of the manuscript for important intellectual content; study supervision

Financial Disclosures: A. Schulman, W. V. Popov - have no personal or financial conflicts of interest to disclose. CC Thompson –Olympus (Consultant/Research Support); Boston Scientific (Consultant); Covidien (Consultant, Royalty, Stock)

No funding

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Albin RL. Sham surgery controls are mitigated trolleys. J Med Ethics. 2005;31:149–52. doi: 10.1136/jme.2003.006155. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Swift T, Huxtable R. The ethics of sham surgery in Parkinson’s disease: back to the future? Bioethics. 2013;27:175–85. doi: 10.1111/j.1467-8519.2011.01931.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Miller FG. Sham surgery: an ethical analysis. Sci Eng Ethics. 2004;10:157–66. doi: 10.1007/s11948-004-0073-x. [DOI] [PubMed] [Google Scholar]
  • 4.Macklin R. The ethical problems with sham surgery in clinical research. N Engl J Med. 1999;341:992–6. doi: 10.1056/NEJM199909233411312. [DOI] [PubMed] [Google Scholar]
  • 5.London AJ, Kadane JB. Placebos that harm: sham surgery controls in clinical trials. Stat Methods Med Res. 2002;11:413–27. doi: 10.1191/0962280202sm300ra. [DOI] [PubMed] [Google Scholar]
  • 6.Oxford English Dictionary. 2. Oxford: Oxford University Press; 1989. [Google Scholar]
  • 7.Higgins J. Cochrane Handbook for systematic reviews of interventions, version 5.1.0 (updated March 2011) 2011. [Google Scholar]
  • 8.Montgomery M, Håkanson B, Ljungqvist O, Ahlman B, Thorell A. Twelve months’ follow-up after treatment with the EndoCinch endoscopic technique for gastro-oesophageal reflux disease: a randomized, placebo-controlled study. Scand J Gastroenterol. 2006;41:1382–9. doi: 10.1080/00365520600735738. [DOI] [PubMed] [Google Scholar]
  • 9.Liu J, Hou X, Song G, Cha H, Yang B, Chen JDZ. Gastric electrical stimulation using endoscopically placed mucosal electrodes reduces food intake in humans. Am J Gastroenterol. 2006;101:798– 803. doi: 10.1111/j.1572-0241.2006.00493.x. [DOI] [PubMed] [Google Scholar]
  • 10.Ponce J, Woodman G, Swain J, Wilson E, English W, Ikramuddin S, et al. The REDUCE pivotal trial: a prospective, randomized controlled pivotal trial of a dual intragastric balloon for the treatment of obesity. Surg Obes Relat Dis. 2015;11:874–81. doi: 10.1016/j.soard.2014.12.006. [DOI] [PubMed] [Google Scholar]
  • 11.Shaheen NJ, Peery AF, Overholt BF, Lightdale CJ, Chak A, Wang KK, et al. Biopsy depth after radiofrequency ablation of dysplastic Barrett’s esophagus. Gastrointest Endosc. 2010;72:490– 496. e1. doi: 10.1016/j.gie.2010.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Toouli J, Roberts-Thomson IC, Kellow J, Dowsett J, Saccone GT, Evans P, et al. Manometry based randomised trial of endoscopic sphincterotomy for sphincter of Oddi dysfunction. Gut. 2000;46:98–102. doi: 10.1136/gut.46.1.98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Fockens P, Cohen L, Edmundowicz SA, Binmoeller K, Rothstein RI, Smith D, et al. Prospective randomized controlled trial of an injectable esophageal prosthesis versus a sham procedure for endoscopic treatment of gastroesophageal reflux disease. Surg Endosc. 2010 Jun 3;24(6):1387– 97. doi: 10.1007/s00464-009-0784-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Bajbouj M, Becker V, Eckel F, Miehlke S, Pech O, Prinz C, et al. Argon plasma coagulation of cervical heterotopic gastric mucosa as an alternative treatment for globus sensations. Gastroenterology. 2009 Aug;137(2):440–4. doi: 10.1053/j.gastro.2009.04.053. [DOI] [PubMed] [Google Scholar]
  • 15.Genco A, Cipriano M, Bacci V, Cuzzolaro M, Materia A, Raparelli L, et al. BioEnterics Intragastric Balloon (BIB): a short-term, double-blind, randomised, controlled, crossover study on weight reduction in morbidly obese patients. Int J Obes (Lond) 2006 Jan 27;30(1):129–33. doi: 10.1038/sj.ijo.0803094. [DOI] [PubMed] [Google Scholar]
  • 16.Devière J, Costamagna G, Neuhaus H, Voderholzer W, Louis H, Tringali A, et al. Nonresorbable copolymer implantation for gastroesophageal reflux disease: a randomized sham-controlled multicenter trial. Gastroenterology. 2005 Mar;128(3):532–40. doi: 10.1053/j.gastro.2004.12.005. [DOI] [PubMed] [Google Scholar]
  • 17.Corley DA, Katz P, Wo JM, Stefan A, Patti M, Rothstein R, et al. Improvement of gastroesophageal reflux symptoms after radiofrequency energy: a randomized, sham-controlled trial. Gastroenterology. 2003 Sep;125(3):668–76. doi: 10.1016/s0016-5085(03)01052-7. [DOI] [PubMed] [Google Scholar]
  • 18.Mathus-Vliegen EMH, van Weeren M, van Eerten PV. Los function and obesity: the impact of untreated obesity, weight loss, and chronic gastric balloon distension. Digestion. 2003;68(2– 3):161–8. doi: 10.1159/000075525. [DOI] [PubMed] [Google Scholar]
  • 19.Clark PA. Placebo surgery for Parkinson’s disease: do the benefits outweigh the risks? J Law Med Ethics. 2002;30(1):58–68. doi: 10.1111/j.1748-720x.2002.tb00720.x. [DOI] [PubMed] [Google Scholar]
  • 20.London AJ, Kadane JB. Sham surgery and genuine standards of care: can the two be reconciled? Am J Bioeth. 2003 Nov;3(4):61–4. doi: 10.1162/152651603322614652. [DOI] [PubMed] [Google Scholar]
  • 21.Horng S, Miller FG. Ethical framework for the use of sham procedures in clinical trials. Crit Care Med. 2003 Mar;31(Supplement):S126–30. doi: 10.1097/01.CCM.0000054906.49187.67. [DOI] [PubMed] [Google Scholar]
  • 22.Johnson AG. Surgery as a placebo. Lancet (London, England) 1994 Oct 22;344(8930):1140–2. doi: 10.1016/s0140-6736(94)90637-8. [DOI] [PubMed] [Google Scholar]
  • 23.Sullivan S, Swain JM, Woodman G, Antonetti M, De La Cruz-Muñoz N, Jonnalagadda SS, et al. Randomized sham-controlled trial evaluating efficacy and safety of endoscopic gastric plication for primary obesity: The ESSENTIAL trial. Obesity (Silver Spring) 2017 Feb;25(2):294–301. doi: 10.1002/oby.21702. [DOI] [PubMed] [Google Scholar]
  • 24.Polgar S, Ng J. Ethics, methodology and the use of placebo controls in surgical trials. Brain Res Bull. 2005 Oct 30;67(4):290–7. doi: 10.1016/j.brainresbull.2005.06.028. [DOI] [PubMed] [Google Scholar]
  • 25.Fregni F, Imamura M, Chien HF, Lew HL, Boggio P, Kaptchuk TJ, et al. Challenges and Recommendations for Placebo Controls in Randomized Trials in Physical and Rehabilitation Medicine. Am J Phys Med Rehabil. 2010 Feb;89(2):160–72. doi: 10.1097/PHM.0b013e3181bc0bbd. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Dowrick AS, Bhandari M. Ethical Issues in the Design of Randomized Trials: To Sham or Not to Sham. J Bone Jt Surgery-American Vol. 2012 Jul 18;94(Suppl 1):7–10. doi: 10.2106/JBJS.L.00298. [DOI] [PubMed] [Google Scholar]
  • 27.BEECHER HK. Surgery as placebo. A quantitative study of bias. JAMA. 1961 Jul 1;176:1102–7. doi: 10.1001/jama.1961.63040260007008. [DOI] [PubMed] [Google Scholar]
  • 28.Kaptchuk TJ, Goldman P, Stone DA, Stason WB. Do medical devices have enhanced placebo effects? J Clin Epidemiol. 2000 Aug;53(8):786–92. doi: 10.1016/s0895-4356(00)00206-7. [DOI] [PubMed] [Google Scholar]
  • 29.Mehta S, Myers TG, Lonner JH, Huffman GR, Sennett BJ. The Ethics of Sham Surgery in Clinical Orthopaedic Research. J Bone Jt Surg. 2007 Jul 1;89(7):1650. doi: 10.2106/JBJS.F.00563. [DOI] [PubMed] [Google Scholar]
  • 30.Gillett GR. Unnecessary holes in the head. IRB. 23(6):1–6. [PubMed] [Google Scholar]
  • 31.Wilcox CM, Lopes TL. A randomized trial comparing endoscopic stenting to a sham procedure for chronic pancreatitis. Clin Trials. 2009 Oct;6(5):455–63. doi: 10.1177/1740774509338230. [DOI] [PubMed] [Google Scholar]
  • 32.Chan MF. Complications of upper gastrointestinal endoscopy. Gastrointest Endosc Clin N Am. 1996 Apr;6(2):287–303. [PubMed] [Google Scholar]
  • 33.Rothstein R, Filipi C, Caca K, Pruitt R, Mergener K, Torquati A, et al. Endoscopic full-thickness plication for the treatment of gastroesophageal reflux disease: A randomized, sham-controlled trial. Gastroenterology. 2006 Sep;131(3):704–12. doi: 10.1053/j.gastro.2006.07.004. [DOI] [PubMed] [Google Scholar]
  • 34.Hartigan PM, Gebhard RL, Gregory PB. Sclerotherapy for actively bleeding esophageal varices in male alcoholics with cirrhosis. Veterans Affairs Cooperative Variceal Sclerotherapy Group. Gastrointest Endosc. 1997 Jul;46(1):1–7. doi: 10.1016/s0016-5107(97)70201-3. [DOI] [PubMed] [Google Scholar]
  • 35.Veterans Affairs Cooperative Variceal Sclerotherapy Group. Prophylactic sclerotherapy for esophageal varices in men with alcoholic liver disease. A randomized, single-blind, multicenter clinical trial. N Engl J Med. 1991 Jun 20;324(25):1779–84. doi: 10.1056/NEJM199106203242505. [DOI] [PubMed] [Google Scholar]
  • 36.Fleischer D. Endoscopic Nd:YAG laser therapy for active esophageal variceal bleeding. A randomized controlled study. Gastrointest Endosc. 1985 Feb;31(1):4–9. doi: 10.1016/s0016-5107(85)71954-2. [DOI] [PubMed] [Google Scholar]
  • 37.Laine L. Multipolar electrocoagulation in the treatment of active upper gastrointestinal tract hemorrhage. A prospective controlled trial. N Engl J Med. 1987 Jun 25;316(26):1613–7. doi: 10.1056/NEJM198706253162601. [DOI] [PubMed] [Google Scholar]
  • 38.Geenen JE, Hogan WJ, Dodds WJ, Toouli J, Venu RP. The Efficacy of Endoscopic Sphincterotomy after Cholecystectomy in Patients with Sphincter-of-Oddi Dysfunction. N Engl J Med. 1989 Jan 12;320(2):82–7. doi: 10.1056/NEJM198901123200203. [DOI] [PubMed] [Google Scholar]
  • 39.Laine L. Multipolar electrocoagulation in the treatment of peptic ulcers with nonbleeding visible vessels. A prospective, controlled trial. Ann Intern Med. 1989 Apr 1;110(7):510–4. doi: 10.7326/0003-4819-110-7-510. [DOI] [PubMed] [Google Scholar]
  • 40.Rutgeerts P, Vantrappen G, Broeckaert L, Coremans G, Janssens J, Hiele M. Comparison of endoscopic polidocanol injection and YAG laser therapy for bleeding peptic ulcers. Lancet (London, England) 1989 May 27;1(8648):1164–7. doi: 10.1016/s0140-6736(89)92751-7. [DOI] [PubMed] [Google Scholar]
  • 41.Mathus-Vliegen EM, Tytgat GN, Veldhuyzen-Offermans EA. Intragastric balloon in the treatment of super-morbid obesity. Double-blind, sham-controlled, crossover evaluation of 500-milliliter balloon. Gastroenterology. 1990 Aug;99(2):362–9. doi: 10.1016/0016-5085(90)91017-z. [DOI] [PubMed] [Google Scholar]
  • 42.Fullarton GM, Birnie GG, Macdonald A, Murray WR. Controlled trial of heater probe treatment in bleeding peptic ulcers. Br J Surg. 1989 Jun;76(6):541–4. doi: 10.1002/bjs.1800760606. [DOI] [PubMed] [Google Scholar]
  • 43.Rutgeerts P, Gevers AM, Hiele M, Broeckaert L, Vantrappen G. Endoscopic injection therapy to prevent rebleeding from peptic ulcers with a protruding vessel: a controlled comparative trial. Gut. 1993 Mar;34(3):348–50. doi: 10.1136/gut.34.3.348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Sclerotherapy for male alcoholic cirrhotic patients who have bled from esophageal varices: results of a randomized, multicenter clinical trial. The Veterans Affairs Cooperative Variceal Sclerotherapy Group. Hepatology. 1994 Sep;20(3):618–25. [PubMed] [Google Scholar]
  • 45.Scolapio JS, Gostout CJ, Schroeder KW, Mahoney DW, Lindor KD. Dysphagia without endoscopically evident disease: to dilate or not? Am J Gastroenterol. 2001 Feb;96(2):327–30. doi: 10.1111/j.1572-0241.2001.03514.x. [DOI] [PubMed] [Google Scholar]
  • 46.Gorelick A, Barnett J, Chey W, Anderson M, Elta G. Botulinum toxin injection after biliary sphincterotomy. Endoscopy. 2004 Feb;36(2):170–3. doi: 10.1055/s-2004-814185. [DOI] [PubMed] [Google Scholar]
  • 47.Mathus-Vliegen EMH, Tytgat GNJ. Intragastric balloon for treatment-resistant obesity: safety, tolerance, and efficacy of 1-year balloon treatment followed by a 1-year balloon-free follow-up. Gastrointest Endosc. 2005 Jan;61(1):19–27. doi: 10.1016/s0016-5107(04)02406-x. [DOI] [PubMed] [Google Scholar]
  • 48.Genco A, Soricelli E, Casella G, Maselli R, Castagneto-Gissey L, Di Lorenzo N, et al. Gastroesophageal reflux disease and Barrett’s esophagus after laparoscopic sleeve gastrectomy: a possible, underestimated long-term complication. Surg Obes Relat Dis. 2016 Dec 9; doi: 10.1016/j.soard.2016.11.029. [DOI] [PubMed] [Google Scholar]
  • 49.Schwartz MP, Wellink H, Gooszen HG, Conchillo JM, Samsom M, Smout AJPM. Endoscopic gastroplication for the treatment of gastro-oesophageal reflux disease: a randomised, shamcontrolled trial. Gut. 2007 Jan 1;56(1):20–8. doi: 10.1136/gut.2006.096842. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Tarnoff M, Rodriguez L, Escalona A, Ramos A, Neto M, Alamo M, et al. Open label, prospective, randomized controlled trial of an endoscopic duodenal-jejunal bypass sleeve versus low calorie diet for pre-operative weight loss in bariatric surgery. Surg Endosc. 2009 Mar 6;23(3):650–6. doi: 10.1007/s00464-008-0125-4. [DOI] [PubMed] [Google Scholar]
  • 51.Aziz AMA, El-Khayat HR, Sadek A, Mattar SG, McNulty G, Kongkam P, et al. A prospective randomized trial of sham, single-dose Stretta, and double-dose Stretta for the treatment of gastroesophageal reflux disease. Surg Endosc. 2010 Apr 3;24(4):818–25. doi: 10.1007/s00464-009-0671-4. [DOI] [PubMed] [Google Scholar]
  • 52.Rodriguez L, Reyes E, Fagalde P, Oltra MS, Saba J, Aylwin CG, et al. Pilot clinical study of an endoscopic, removable duodenal-jejunal bypass liner for the treatment of type 2 diabetes. Diabetes Technol Ther. 2009 Nov;11(11):725–32. doi: 10.1089/dia.2009.0063. [DOI] [PubMed] [Google Scholar]
  • 53.Gersin KS, Rothstein RI, Rosenthal RJ, Stefanidis D, Deal SE, Kuwada TS, et al. Open-label, shamcontrolled trial of an endoscopic duodenojejunal bypass liner for preoperative weight loss in bariatric surgery candidates. Gastrointest Endosc. 2010 May;71(6):976–82. doi: 10.1016/j.gie.2009.11.051. [DOI] [PubMed] [Google Scholar]
  • 54.Arts J, Bisschops R, Blondeau K, Farré R, Vos R, Holvoet L, et al. A double-blind sham-controlled study of the effect of radiofrequency energy on symptoms and distensibility of the gastroesophageal junction in GERD. Am J Gastroenterol. 2012 Feb 22;107(2):222–30. doi: 10.1038/ajg.2011.395. [DOI] [PubMed] [Google Scholar]
  • 55.Mathus-Vliegen EMH, de Groot GH. Fasting and meal-induced CCK and PP secretion following intragastric balloon treatment for obesity. Obes Surg. 2013 May 8;23(5):622–33. doi: 10.1007/s11695-012-0834-6. [DOI] [PubMed] [Google Scholar]
  • 56.Thompson CC, Chand B, Chen YK, Demarco DC, Miller L, Schweitzer M, et al. Endoscopic suturing for transoral outlet reduction increases weight loss after Roux-en-Y gastric bypass surgery. Gastroenterology. 2013 Jul;145(1):129–137. e3. doi: 10.1053/j.gastro.2013.04.002. [DOI] [PubMed] [Google Scholar]
  • 57.Mathus-Vliegen EMH, Eichenberger RI. Fasting and meal-suppressed ghrelin levels before and after intragastric balloons and balloon-induced weight loss. Obes Surg. 2014 Jan 6;24(1):85–94. doi: 10.1007/s11695-013-1053-5. [DOI] [PubMed] [Google Scholar]
  • 58.Eid GM, McCloskey CA, Eagleton JK, Lee LB, Courcoulas AP. StomaphyX vs a sham procedure for revisional surgery to reduce regained weight in Roux-en-Y gastric bypass patients: a randomized clinical trial. JAMA Surg. 2014 Apr 1;149(4):372–9. doi: 10.1001/jamasurg.2013.4051. [DOI] [PubMed] [Google Scholar]
  • 59.Cotton PB, Durkalski V, Romagnuolo J, Pauls Q, Fogel E, Tarnasky P, et al. Effect of endoscopic sphincterotomy for suspected sphincter of Oddi dysfunction on pain-related disability following cholecystectomy: the EPISOD randomized clinical trial. JAMA. 2014 May 28;311(20):2101–9. doi: 10.1001/jama.2014.5220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Håkansson B, Montgomery M, Cadiere GB, Rajan A, Bruley des Varannes S, Lerhun M, et al. Randomised clinical trial: transoral incisionless fundoplication vs. sham intervention to control chronic GERD. Aliment Pharmacol Ther. 2015;42:1261–70. doi: 10.1111/apt.13427. [DOI] [PubMed] [Google Scholar]

RESOURCES