Diclophenac Versus Placebo for Pain Control in Diagnostic Colonoscopy

Aim:

To determine the effectiveness of oral diclophenac sodium on the quality of colonoscopy and pain control during diagnostic colonoscopy in hospitalized patients.

Hypothesis:

Null hypothesis: there is no statistically significant difference between diclophenac sodium and placebo on parameters of quality of colonoscopy and pain control in hospitalized patients undergoing diagnostic colonoscopy.

Outcomes:

Primary outcomes:

Reduction of total mean pain index by 2 points (as measured on a 10-point Likert scale)

Secondary outcomes:

Reduction of pain in the subgroup of patients with high pain index (defined as patients that report 7-10, inclusive, on a 10-point Likert scale) Reduction of pain in the subgroup of patients with moderate pain index (defined as patients that report 3-6, inclusive, on a 10-point Likert scale) Proportion of patients that are willing to repeat the procedure, if necessary Need for sedation or analgesia during colonoscopy Time to discharge

Adverse events:

Total complications (as recorded from the beginning of colonoscopy to hospital discharge) Colonoscopy-related complications (defined as any vasovagal, hypoxemic, bleeding or perforation event) Total mortality (defined as any lethal outcome from the beginning of colonoscopy to hospital discharge) Patients will be followed until hospital discharge, an average of 10 days. Sample size calculation To be able to detect a 2-point difference in mean pain index with standard deviation of 3, with alpha=0.05 and power=0.80, the investigators calculated the required total sample size of 72 patients.

Methods This is a single-center, prospective, randomized, placebo-controlled, double-blind clinical trial of diclophenac sodium versus placebo for colonoscopy. The investigators plan to enroll a total of 72 patients that are referred for diagnostic colonoscopy for any indication during their hospital stay. The trial will be conducted in a university hospital in Croatia (University hospital Rijeka, Department of Gastroenterology).

Inclusion criteria: all inpatients that are referred for colonoscopy for any indication during their hospital stay.

Exclusion criteria:

refusal to sign informed consent age <18 years pregnancy lactation allergy to any nonsteroidal antiinflammatory drugs patients that have taken any nonsteroidal antiinflammatory drug in the previous seven days before randomization (except acetylsalicylic acid in doses up to 300 mg/day) patients on antiplatelet or anticoagulation therapy (except acetylsalicylic acid in doses up to 300 mg/day) history of gastric or duodenal ulcer history of GI bleeding or perforation history of or active inflammatory bowel disease severe liver disease (defined as presence of history of ascites and/or esophageal varices) severe kidney disease (defined as glomerular filtration rate <30 ml/min) history of myocardial infarction or cerebrovascular disease history of peripheral arterial disease congestive heart failure (NYHA III-IV) history of systemic lupus erythematosus Protocol: Patients referred for colonoscopy during their hospital stay will be informed about the trial protocol and possible risks and adverse events, after which they will be offered to accept to be included in the trial by signing the informed consent form. Upon inclusion, patients will be randomized by computer random number sequence generation to either 'Group 1' (diclophenac) or 'Group 2' (placebo). Patient allocation will be withheld from all participants in the trial and stored at a secure central location.

Group 1 will receive peroral diclophenac sodium in the dose of 100mg (two 50mg tablets), two hours before colonoscopy, on an empty stomach with one glass of water. Diclophenac sodium tablets will be provided by the hospital pharmacy.

Group 2 will receive two tablets of placebo that are same shape, size, colour and in no other way different from diclophenac tablets. Placebo will be provided by a collaborating pharmacy.

The patients, attending physicians, endoscopists, data assessors and data analysts will all be blinded to the patients' allocation. Allocation will be kept at a central location and will be known only to the person who will randomize patients. On the day of colonoscopy all patients will be accompanied with a trial form where all relevant data will be entered. First part will be filled by the attending physician who refers the patient to colonoscopy, the second part will be filled by the endoscopist and the third part will be filled by trial associates on the day of discharge (appendix 1).

All patients will be assessed before randomization by history and physical examination, laboratory tests that will include complete blood count, plasma glucose, urea, creatinine, glomerular filtration rate, Na, K, total and direct bilirubin, aspartate transaminase, alanine transaminase, alkaline phosphatase, gamma-glutamyltransferase, albumin, serum protein electrophoresis, coagulogram (prothrombin time and INR), urinalysis, chest radiogram and electrocardiogram.

Statistical analysis The investigators will use two-tailed Wilcoxon-Mann-Whitney test for means and two-tailed Fisher's exact test for proportions. The investigators will use Kaplan-Meier curves for construction of time to discharge graphs and Mann-Whitney U test for statistical difference between the curves. The investigators will perform multivariate analysis of different characteristics to test for effect of those characteristics on the result of mean pain index as reported by the patient. If any of the characteristic shows statistically significant effect, the investigators will perform univariate analysis on those covariates.

Financial support This trial will be supported from the Department of Gastroenterology's own resources, considering that the drug investigated is routinely administered to patients with indications.