Two hundred and twenty five women with post menopausal bleeding will be randomly divided into three equal groups. To ensure blinding the investigators will use the double dummy technique in which group 1 will receive Tramadol 100mg (Trama®, Global Napi, Giza, Egypt) orally in addition to a placebo similar to Celecoxib, group 2 will receive Celecoxib 200mg (Celebrex® 200, Pfizer, USA) in addition to a placebo similar to Tramadol, and group 3 will received a placebo similar to Tramadol and a placebo similar to Celecoxib. All drugs will be given 2 hours before the procedure. An independent person will generate the allocation sequence using computer generated random numbers.
The procedure will be done in the lithotomy position. The investigators will use a 30 degree angle 2.7 mm rigid hysteroscope with a 5mm outer diameter continuous flow hysteroscope with a 5 French working channel, a Teknolight 180 XA light source, and a Xenon high density fibre optic light cable 3.5mm, 2300mm Light cable without adaptors and a T Camera teknocam 2000S pro. All the equipment are provided by Tekno GmbH and Co®, Germany.
Vaginoscopic approach will be used for insertion of the hysteroscope in all cases (no use of speculum or tenaculum). The hysteroscope will be gently introduced into the uterine cavity after visualization of the cervix and identification of the external os. The investigators will use saline as the distension medium and the maximum pressure will be set at 80mm Hg. The uterine cavity and tubal ostia will be systematically visualized. Endometrial biopsy will be taken using a semi-rigid, double action oval serrated biopsy forceps 40cm 5Charr provided by Tekno GmbH and Co®. The most suspicious area will be identified, the forceps will be advanced and opened against the suspicious area and the tissue will be grasped, rotated 90 degrees and removed with the hysteroscope out of the cervical canal Patient's perception of pain will be assessed for each group during the procedure, immediately after and 30 min after the procedure with the use of visual analogue scale (VAS). VAS of 0 indicates no pain and VAS of 10 indicates the worst possible experienced pain. Patients will also be asked to report any side effects. The main outcome measure will be patients' pain perception during the procedure and the secondary outcome will be the adequacy of the sample.
Statistics:
Quantitative data will be statistically represented in terms of mean ± standard deviation (± SD) while categorical data will be represented as frequency and percentage. Comparison of quantitative data will be done using ANOVA test for independent samples while categorical data will be compared using Chi squared test or Fisher exact test when appropriate. A probability value (p value) less than 0.05 will be considered significant.
Sample size calculation:
To the best of the investigators' knowledge this is the first trial to investigate the role of Celecoxib and oral Tramadol in reducing outpatient hysteroscopy associated pain in post menopausal women, with no previous data to calculate the sample size with in this cohort of women. Data from studies on premenopausal women cannot be applied to post menopausal women because hysteroscopies in post menopausal women are usually more difficult, require more cervical manipulations and cause more pain. Assuming that the response will be normally distributed, the sample size is calculated to detect a mean difference of 1 unit between Tramadol and Celecoxib pain scores during the procedure (lower difference are not considered clinically relevant) using VAS and assuming that the within group standard deviation will be 2. The investigators will need to study 64 cases in each group to be able to reject the null hypothesis that the population means of the Tramadol and Celecoxib are equal with probability (power) 0.8. The investigators added 11 cases to each arm accounting for any missing data and procedure failure ending in 75 cases in each group. The Type I error probability associated with this test of this null hypothesis is 0.05 using Student's t test for independent samples. Sample size calculation is done using Stats Direct statistical software version 2.7.2 for MS Windows, Stats Direct Ltd., Cheshire, UK.
