Comparison of Fluid Capture Efficiency of the Total Capture Drape Versus a Standard Drape for Hysteroscopy

EXPERIMENTAL TRIAL WITH THE TCD VERSUS THE STANDARD HYSTEROSCOPY DRAPE IN A PELVIC MODEL SIMULATION

The investigators sought to document whether or not a prototype design of the "Total Capture" drape (TCD) would perform under replicated clinical conditions in a superior fashion compared to the currently marketed hysteroscopy drape. The experimental testing was performed at an operating room suite using the hysteroscopy fluid management system, conventional hysteroscopy drape, the Total Capture drape and the ZOE Gynecologic pelvic model. The pelvic model was configured so that all fluid introduced into the artificial vagina would not be absorbed by the model but would flow out of the artificial vagina.

A total of four trials per drape were conducted. One trial from each product had procedural difficulties making results invalid.

Results of the Simulation Trial: In four trials of the total capture drape, there was no deficit observed. It was concluded that no major design iterations would be required to move the TCD prototype to the next phase of development

COMPARATIVE TRIAL FOR "PROOF OF CONCEPT" TESTING TO EVALUATE THE EFFICIENCY OF THE TOTAL CAPTURE DRAPE COMPARED TO THE CONVENTIONAL HYSTEROSCOPY DRAPE DURING HYSTEROSCOPIC SURGERY

All women undertaking hysteroscopy are required to have a sterile surgical drape with a fluid capture pouch system to collect the distention media not absorbed by the patient to calculate fluid deficits. Both the TCD and the standard hysteroscopy drape possess the characteristics described above, although design features of the standard drape are deficient. The TCD falls under Class II status, exempt from the 510(k) clearance process in consideration of FDA's approach to similar devices like the standard hysteroscopy drape (Regulation number 21 CRF 878.4370, product code KKX.

The hysteroscopy fluid deficit will be measured by the fluid management system during each hysteroscopy procedure. The uncollected fluid volume seen on the operating room floor and absorbed by the operation blankets and drapes will also be captured, as well as the estimated fluid loss from leaking hysteroscopy joints and gaskets not captured in the collection bag. Procedure time and alarm sounding for exceeding fluid deficit limit will also be recorded.

MEASURES OF DRAPE USABILITY, QUALITY, AND CONVENIENCE

Staff will be asked to complete the System Usability Scale (SUS) to evaluate their perceived clinical usefulness of the TCD and the standard drape. The SUS is a validated 10-item survey for assessing perceived usability that uses a 5-point Likert scale ranging from strongly disagree to strongly agree. To further subjectively and objectively assess the TCD, a human factors and design researcher will observe the surgeries. The human factors researcher will attend the first surgeries to ensure consistent rating with the design researcher who will subsequently observe the remaining surgeries until data saturation is reached. Staff will be asked to speak aloud using the system, which the researcher will record. Further, the researcher will complete an objective human factors assessment tool. These metrics will allow us to quantitate clinical usefulness and usability of both the operating surgeons and operating room staff. These results will not only provide this assessment of the TCD compared to the standard hysteroscopy drape (which is used across ORs within Atrium Health) but will also provide important information to improve the drape if deficiencies are identified.

DATA MANAGEMENT

Case report forms will be developed by the OBGYN clinical research nursing staff in collaboration with Dr. Marshburn and Dr. Zhao. A REDCap database will be created for secure electronic data capture. The data flow will consist of paper data collection for eligibility assessment, baseline data, randomization, operational procedure, fluid measures, surgery outcomes, adverse events, and protocol deviations. The data will be entered into the electronic database within 3 business days of the enrollment.

STATISTICAL METHODS AND SAMPLE SIZE JUSTIFICATION

The primary outcome for the study is the hysteroscopy fluid deficit (HFD) as determined by the fluid management system by subtracting the outflow volume (OV) from the inflow volume (IV) (HFD=IV-OV). Our null hypothesis is Ho: μStandard=μTCD where μ=mean fluid deficit. Our alternative hypothesis is Ho: μStandard>μTCD due to the underestimation of the outflow volume created by the loss of outflow fluid to floors and blankets. If it is assumed the true absorption rates by the patients are the same between the groups due to randomization, by estimating and testing the difference HFD between the two groups a more accurate measure of the uncollected fluid volume (UFV) will be estimated and tested as shown below:

μStandard - μTCD = (μStandard, true absorption + μStandard, UFV) - μTCD, true absorption

• (μ true absorption + μStandard, UFV) - μtrue absorption
• μStandard, UFV The primary analysis for the study will use a general linear model on the HFD with a fixed effect for study group (TCD versus Standard drape) controlling for surgeon and indication for hysteroscopy, the stratification factors for the randomization.

Secondary outcomes include estimated uncollected fluid volume from floor and blankets, occurrence of the fluid deficit alarm sounding, stopping of the operation due to concerns of overabsorption by the patient, and adverse events. Estimated uncollected fluid volume will be analyzed similar to the primary outcome. All other outcomes are categorical and will be compared using Chi-square analyses.

The investigators will sample 10 procedures(or until data saturation) from each arm (TCD vs standard) to conduct qualitative assessments on Usability using the Human Factors Assessment Tool /Acceptance for the for the surgeons and operating team members.For the SUS, the design strategist will collect data on attending physicians only on their first and fifth procedures (the study team may change it to 1st and 3rd if deemed appropriate). The investigators will estimate means or medians with standard deviations (or interquartile ranges) for the items and total scores for the measuresSUS. Both within (1st vs. 5th) and between arms (TCD vs. standard) comparison on total scores of SUS will be conducted. The investigators will compare the two groups using linear mixed models or generalized linear mixed models (depending on the distribution) with a random effect for the procedure to control for correlation of the measures among team members participating in the same procedure.

There is no prior data on HFD in patients to estimate the variability need for the sample size estimation. There is preliminary data from the simulated experiment showing a 100% recovering of loss deficit by the TCD. If it is conservatively assumed the loss deficit will decrease from 40% to 20% of the inflow volume, this would result in a clinically meaningful improvement with an approximate effect size of 0.92 based on the variability of loss deficit in the simulation. Given this, effect size (relative to standard deviation) is used as the primary justification for the number of patients needed for the study. With a sample size of n=34 per arm, there would be 81% power to detect an effect size of 0.7 (0.7*standard deviation of the HFD) with a two sided, two sample t-test (α=0.05). To account for potential missing data of 10%, it is proposed randomizing 76 women (n=38 per arm). With n=34 women in each group, there would also be 80% power to detect an absolute 33% reduction in the proportion of procedures with the alarm sounding due to fluid deficit threshold being reached assuming the proportion in the standard drape arm is 50%. This power was calculated using a Fisher's Exact test for testing proportions between two independent samples. The same power would apply for detecting differences in the proportion of procedures terminated.

DATA SAFETY AND MONITORING

The PI and the study team will meet biweekly to review the study procedures, enrollment (screening and randomization), implementation, protocol deviations, data collection, and adverse events. The biostatistical team will generate monthly reports to monitor screening, enrollment, completeness of data for intervention implementation and outcomes, adverse events, and protocol deviations.