Intrauterine Pressure During Hysteroscopic Morcellation: A ... · An endoscopic optical lens is inserted through the cervix into the endometrial cavity to directly visu - alize and

SStudy objective: Our objective was to compare intrauterine pressures during resection and aspiration modes

among three types of commercially-available hysteroscopic morcellators.

Design: This was a benchtop study (Canadian Task Force level II-1). This study cannot feasibly and ethically be

done in-vivo, so an ex-vivo study design was chosen.

Setting: A silicone uterine model was attached to a manometer via tubing, with the tip inside the cavity to allow

for intracavity pressure measurements. Each hysteroscopic morcellator was then introduced, and intracavity

pressures were recorded every one to two seconds in three modes (static, resection, and aspiration) and at three

set point pressures (45, 85, and 125 mmHg).

Patients: No human subjects were involved in this study.

Interventions: None.

Measurements and main results: There were a total of 4,872 pressure measurements during this study across the

three devices, over the three modes, and at the three set point pressures combined. Using mixed-effects linear

regression, the mean observed intracavity pressure was not greater than the set pressure for each of the three

devices. This result held true in both aspiration and resection modes. In our statistical models, the coefficient

on the terms representing the interaction between device and time were not statistically significant in either

resection or aspiration modes. This indicates that, statistically, the change in intracavity pressure over time was

not significantly different across the three devices.

Intrauterine Pressure During HysteroscopicMorcellation: A Comparison of Three

Commercially-Available Devices

- 1 -

#1048 Howard FINAL

ABSTRACT

GynecologySURGICAL TECHNOLOGY INTERNATIONAL Volume 33

ERICA STOCKWELL, DO, MBADIRECTOR OF INNOVATION AND BUSINESS EDUCATIONMINIMALLY INVASIVE GYNECOLOGIC SURGEONLAS VEGAS MINIMALLY INVASIVE SURGERY

WELL HEALTH QUALITY CAREA DAVITA MEDICAL GROUP

CLINICAL ASSISTANT PROFESSORDEPARTMENT OF OBSTETRICS AND GYNECOLOGY

UNIVERSITY OF NEVADA, LAS VEGAS SCHOOL OF MEDICINELAS VEGAS, NEVADA

DAVID L. HOWARD, MD, PHDDIRECTOR OF RESEARCH

LAS VEGAS MINIMALLY INVASIVE SURGERYWOMEN'S PELVIC HEALTH CENTER/WELLHEALTH QUALITY CAREA DAVITA MEDICAL GROUP

CLINICAL ASSISTANT PROFESSORDEPARTMENT OF OBSTETRICS AND GYNECOLOGY

UNIVERSITY OF NEVADA, LAS VEGAS SCHOOL OF MEDICINELAS VEGAS, NEVADA

- 2 -

Hysteroscopy is a commonly-usedminimally invasive gynecological proce-dure, utilized in both clinical and oper-

ating settings. An endoscopic opticallens is inserted through the cervix intothe endometrial cavity to directly visu-alize and treat pathology. Operative hys-teroscopy became popular afterimprovements in endoscopic technology

and instruments in the 1970s and afterthe introduction of the fluid distensionmedial in the 1980s. Since that time, thedevelopment of new hysteroscopicinstruments, fiber optics, and digitalvideo equipment have to provide more

#1048 Howard FINAL

Intrauterine Pressure During Hysteroscopic Morcellation: A Comparison of Three Commercially-Available DevicesSTOCKWELL/HOWARD

INTRODUCTION

Conclusion: In this first of its kind head-to-head benchtop study, we found that all three commercially-

available hysteroscopic morcellators appear to be similar to each other in terms of their abilities to maintain

intracavity pressure below the set pressure, which is important in avoiding intravasation in-vivo. These findings

are important because many gynecologists do not have the ability to choose between the three available devices

on the market at their institution.

Table I Descriptions of Symphion™, TruClear™ and MyoSure® hysteroscopic

morcellator systems

Type of energyduring resection

Components Description from manufacturer

Symphion™

Boston Scientific

Bipolar radiofrequency

Symphion™ controllerwith integrated fluidmanagement

2 modes: diagnostic and resection. Figure 1a

Symphion™ fluid man-agement Accessories, includingfootswitch and salinepole

Fluid infusion and aspiration of the uterine cavity arecontrolled by the controller’s peristaltic pumps, in con-junction with the disposable fluid management acces-sories; these components form a closed-loopre-circulating system.

Figure 1a

Symphion™ resectingdevice

Disposable hand-held bipolar radiofrequency device. Figure 1b

Symphion™ endoscope 0° 6.3 mm rigid scope, with 2 integrated fluid channels(1.5 mm) and 1 working channel (3.7 mm). The fluidchannels are used for infusion or distension fluid anddirect pressure monitoring of the cavity. The workingchannel accommodates the SymphionTM resectingdevice. The working length of the SymphionTM endo-scope is 208 mm.

Figure 1c

Truclear™

Medtronic(Smith &Nephew)

Mechanical TruClear™ fluid management accessories, includingfootswitch

Minimizes fluid use through proprietary suction controltechnology. Provides simultaneous cutting and tissue removal,requiring only a single insertion.Provides fewer procedural steps due to single insertion.

Figure 2a

TruClear™ tissueremoval devices

1. TruClear™ INCISOR™ device for soft tissue 2. TruClear™ INCISOR™ Plus device for soft tissue 3. TruClear™ ULTRA Mini device for dense tissue 4. TruClear™ ULTRA Plus device for dense tissue

Figure 2b

TruClear™

hysteroscope1. TruClear™ 5C hysteroscope set-Scope is 5 mm in diameter with a 5.7 mm sheath-0° rigid scope

2. TruClear™ 8.0 hysteroscope set-0° rigid scope-For use with larger tissue removal devices (INCISOR™ Plus and ULTRA Plus)

Figure 2c

- 3 -

#1048 Howard FINAL


Table I (continued)Descriptions of Symphion™, TruClear™ and MyoSure® hysteroscopic

morcellator systems

MyoSure®

HologicMechanical MyoSure® hystero-

scope0° direction of view and 80° field of view.Small 6.25 mm outer diameter.Larger scope available that is 7.25mm diameter.Outflow channel is removable.

Figure 3

Tissue removaldevices

MyoSure® LITE: Window length: 10.2 mmWindow depth: 1.5 mm

MyoSure®: REACHWindow length: 14 mmWindow depth: 1.8 mm

MyoSure® XLWindow length: 14 mmWindow depth: 2.4 mm

Symphion figures and description obtained from:https://www.bostonscientific.com/content/dam/bostonscientific/uro-wh/sites/symphion/pdfs/System_Instructions.pdf.Truclear figures and description obtained from:http://www.premiermedical.ie/wp-content/uploads/2013/09/TRUCLEAR-Brochure.pdf.Myosure figures and description obtained from: http://myosure.com/sites/myosure/files/PB-00280-001_Rev_002_MS_Phy_Brochure.pdf.

Figure 1b. Symphion™ resecting device. Figure 1c. Symphion™ endoscope.

Figure 1a. Symphion™ controller with integrated fluid management.

- 4 -

varied, efficacious, and less invasiveprocedures.1Operative hysteroscopy is, overall, a

safe procedure, resulting in complica-tions in 0.95–3% of cases.2-4 The mostfrequently-observed complicationsinclude hemorrhage (2.4%), uterineperforation (1.5%), and cervical lacera-tion (1–11%).5 Another rare complica-tion is excessive fluid absorption, withor without resulting hyponatremia(0.2–0.76%).3,4,6 Fluid deficits shouldbe carefully managed during hys-teroscopy to prevent intravasation.Lower uterine filling pressures havebeen associated with lower patient painscores but a higher trend toward inade-quate visibility.7The purpose of this non-human

benchtop study was to compare intra-cavity pressures during resection andaspiration modes among three commer-cially-available hysteroscopic morcella-tors, using a model to simulate thehuman uterus. No prior study has com-pared these three devices in a head-to-head manner using a clinically-relevantoutcome measure. Although ex-vivostudies are sometimes limited in gener-alizability, some studies cannot feasiblyor ethically be carried out in-vivo.

#1048 Howard FINAL


Figure 2a. TruClear™ fluid management acces-sories, including footswitch.

Figure 2b. TruClear™ tissue removal devices.

Figure 2c. TruClear™ hysteroscope.

Figure 3. MyoSure® hysteroscope.

- 5 -

There is no feasible way to obtain con-tinuous intrauterine pressure measure-ments in a live patient. It is, however,very important to attempt to directlycompare the three commercially-avail-able hysteroscopic morcellators interms of their abilities to maintainintrauterine pressure below the setpressure.

Materials and Methods

Intrauterine pressure and accuracywas compared utilizing Symphion™(Boston Scientific Corporation, Marl-borough, Massachusetts), MyoSure®(Hologic, Inc., Marlborough, Massachu-setts), and TruClear™ (Medtronic plc.,Fridley, Minnesota) systems in threemodes (static, aspiration, and resection)and at different pressure settings. Thestudy was performed using a siliconeuterine model, a manometer to mea-sure intrauterine pressure at one-secondintervals, and porcine heart tissue torepresent intrauterine fibroid material.Each arm was performed three times(to account for possible variabilitybetween individual devices), each timewith a new device, and results wereaveraged. In Table I and Figures 1–3, weprovide a detailed description of each ofthe three devices.

Pressure control setupFirst, a 20-gauge dispensing tip

(manufactured by Nordson EFD) wasused to pierce the thickened section ofthe simulated uterine cavity model (Fig.4). Next, the endoscope was introducedinto the clear cavity model without theresection device attached. Then, themanometer was connected to the dis-pensing tip and was connected to acomputer using the USB cable providedwith the manometer. This allowed formeasurement of actual pressure in thecavity model, which was compared tothe set pressure on the controller’sgraphical user interface (GUI). Sper Sci-entific Data Acquisition Software (SperScientific Ltd., Scottsdale, Arizona) wasused to measure intrauterine pressure.

Static pressure controlThe cavity pressure on the controller

was set to 45 mmHg. An infusion pumpwas run, and cavity pressure was mea-sured after 30 seconds. Cavity pressurewas increased in 10 mmHg increments

every 30 seconds until 125 mmHg wasreached. Pressure was measured at one-second intervals during each step. Thesesteps were performed equally for allthree device systems. The primary pur-pose of this paper was to evaluate theintrauterine pressures observed duringaspiration and resection, which aredescribed in the following sections.

Aspiration pressure controlThe cavity pressure on the controller

was set to 45 mmHg. An infusion pumpwas run, and aspiration was performedfor 10 seconds. Cavity pressure wasmeasured at one-second intervalsthroughout. This was repeated at cavitypressures of 85 and 125 mmHg for allthree device systems.

Resection pressure controlPorcine heart tissue was used to sim-

ulate human uterine fibroid tissue. Two1.5-inch portions of tissue were insert-ed into the silicone model, and theendoscope with resection device wasinserted. Cavity pressure was set to 45mmHg at the controller. Cut mode wasactivated for 10 seconds, and cavitypressure was measured at one-secondintervals throughout. This was repeatedat set cavity pressures of 85 and 125mmHg for all three device systems.

Statistical methodologyFor each device, there were three

trials. During each trial, the set pointpressure was systematically increasedfrom one predetermined level toanother. Intrauterine pressure measure-ments were repeatedly taken, at everysecond, for each given set point pres-sure. Therefore, the data consisted of

“repeated measures” and, as such, mul-tilevel modeling was used to analyzethe data.In constructing our mixed-effects

multilevel model, we specified dummyvariables representing the device as thefixed effect. Symphion™ was set to bethe reference category for each of thetwo dummy variables representingMyoSure® and TruClear™. In additionto dummy variables representing thedevice, we also included an interactionterm as a covariate, representing theinteraction between device and time.The purpose of this interaction termwas to be able to assess whether thevariation in intrauterine pressure overtime was different for the three devices.The stratification variable used wasmode. We conducted a separate modelfor aspiration and resection modes.The dependent variable in our mod-

els was the measured intrauterine pres-sure. We centered the observedintrauterine pressure reading by sub-tracting the set point pressure from theobtained value. For example, if theobserved intrauterine pressure was 100mmHg and the set point pressure was85 mmHg, then the centered pressurewould be 15 mmHg (100–85 mmHg).We did this to allow the dependent vari-able to be continuous, thereby main-taining statistical power and allowingthe interpretation of the results fromthe model to be more clinically mean-ingful. In our linear mixed models, ourmain clinical concern was whether thepredicted mean intrauterine pressureduring use of each device was signifi-cantly higher than the set pressure, asthis would put the patient at higher riskfor intravasation.

#1048 Howard FINAL


Figure 4. Twenty-gauge Nordson EFD dispensing tip.

MATERIALS AND METHODS

- 6 -

Finally, we used graphical methodsto construct two sets of plots. First, weconstructed plots of centered intrauter-ine pressure against time in seconds foreach device in each mode. In the secondset of plots, we plotted centered

intrauterine pressure against set pres-sure for each device, irrespective ofmode.We used the mixed command in

STATA (StataCorp LLC., College Sta-tion, Texas; Version 14) to do the above

models and the lowess command toobtain smoothed plots. Our detailedstatistical codes are available uponrequest.

Results

In Table I and Figures 1–3, we pre-sent a descriptive comparison of thethree commercially-available hystero-scopic devices. One of the sentinel dif-ferences is that one device (Symphion™)uses bipolar energy that allows for coag-ulation during resection and aspiration,while the other two devices utilize sim-ple mechanical energy.

Mean intrauterine pressure acrossthe three devicesThere were a total of 4,872 pressure

recordings in this experiment. The pre-dicted mean-centered intracavity pres-sures obtained from our linear mixed

#1048 Howard FINAL


Table IIAssociation between intrauterine pressure and

type of device based on mixed effects multi-level regression modeling

Aspiration Resection

Predicted cen-tered intrauterine

pressurea95% CI

Predicted cen-tered intrauter-ine pressurea

95%CI

DeviceMyosure®

Symphion™

TruClear™

-14.2-18.4-9.4

-30.9, 2.5-20.7, -16.0-17.1, -1.7

-12.4-18.3-18.9

-23.4, -1.4-23.3, -13.3-23.9, -13.9

a Obtained from models that were constructed as linear mixed effects multi-levelmodels. Dependent variable was the centered intrauterine pressure in mmHg.

Figure 5. Centered intrauterine pressure versus time in static, resection, and aspiration modes for three hysteroscopic morcellators.

RESULTS

- 7 -

models are shown in Table II. The clini-cally-meaningful outcome is whetherthe mean intrauter ine pressure isgreater than the set pressure. In this sce-nario, the patient would be at risk forintravasation of fluids. In this study, inaspiration mode, the mean-centeredintracavity pressure was never higherthan 0, regardless of which device wasused. In other words, for each of thethree devices, the mean observed intra-cavity pressure was not greater than theset pressure. The confidence intervalsaround the mean-centered intracavitypressure for each device all overlap,indicating that the difference betweenthe mean intracavity pressure and theset pressure was not statistically differ-ent across the three devices.In resection mode, the same pattern

was observed. The predicted mean-cen-tered intracavity pressure was nevergreater than 0, regardless of whichdevice was studied. Also, the confidenceintervals around the mean-centered

intracavity pressure for each device alloverlapped with each other.It is worth noting that, in both resec-

tion and aspiration modes, the confi-dence interval around the meanintracavity pressure for MyoSure®appeared to be much wider than theconfidence interval around the meanintrauterine pressure for Symphion™and TruClear™.

Variation of intrauterine pressureover timeIn Figure 5, we show the variation in

intrauterine pressure over time for eachdevice in each mode. In both aspirationand resection modes, the variation inintrauterine pressure over time graphi-cally appears to be lower forSymphion™, compared to the other twodevices (MyoSure® and TruClear™).However, in our statistical models, thecoefficient on the terms representingthe interaction between device and timewere not statistically significant in either

resection or aspiration modes. Thisindicates that, statistically, the change inintrauterine pressure over time was notsignificantly different across the threedevices.

Variation of intrauterine pressureby set pressureIn Figure 6, we show the variation of

intrauterine pressure by set pressure(for all modes combined). In aspirationand resection modes, only three setpressures were used (45, 85, and 125mmHg), so, not surprisingly, more vari-ability was observed at these three setpressures. However, the variation inintrauterine pressures at these three setpressures, graphically speaking,appeared to be lower for Symphion™,compared to the other two competitordevices (MyoSure® and TruClear™). Wedid not formally test whether the varia-tion in the intrauterine pressure at eachset pressure was statistically differentacross the three devices.

#1048 Howard FINAL


Figure 6. Variation of intrauterine pressure by set pressure for all modes combined for three hysteroscopic morcellators.

- 8 -

Discussion

We present a unique ex-vivo bench-top study of three commercially-avail-able hysteroscopic morcellating devices.For all three devices, regardless ofwhether the mode was aspiration orresection, the mean measured intracavi-ty pressure was below the set pressure.This provides some support to the basicconclusion that all three devices areequally safe in terms of the risk ofintravasation. This is the first head-to-head benchtop

study comparing the three commercially-available hysteroscopic morcellators inthe United States. We believe this to be avery well-thought-out experiment with alarge number of data points (4,872), andwe believe our statistical methods wererigorous. The main limitation with this study

was that it was ex-vivo and not in-vivo.It is unknown how much the results ofan experiment conducted using a sili-cone model to simulate the uterus willtranslate to a real hysteroscopic proce-dure on a real patient. However, someexperiments are very difficult to exe-cute in-vivo and, consequently, ex-vivostudies such as ours still play an impor-tant role in informing clinical practice.It would be extremely technically chal-lenging to conduct an in-vivo study thatinvolves continuous intrauterine pres-sure measurement while a hysteroscop-ic procedure is used.Our main unanswered question is

whether variation in intrauterine pres-sure over time has clinical relevance.Although the change in pressure overtime was not statistically differentacross the three devices, based on statis-tical models, we noticed that the confi-dence interval around the meanintrauterine pressure was much widerfor MyoSure®, compared to TruClear™

and Symphion™. In aspiration mode, forexample, the width of the confidenceinterval was 33.4 mmHg, whereas forSymphion™, it was only 4.7 mmHg. Webelieve that future studies shouldexplore whether there is any clinicalrelevance to variations in intrauterinepressure over time during hysteroscopicmorcellation. Another observation that deserves

further exploration is the fact that, inaspiration mode, the entire confidenceinterval around the mean intrauterinepressure for the Symphion™ device wasbelow the mean intrauterine pressurefor the other two devices. Does thisindicate a potential tendency for Sym-phion™ to “under-pressurize” the uteruscompared to the other two devices?And, if so, does this have clinical rele-vance? Further studies are needed tomore carefully answer these questions.

Conclusion

In conclusion, this detailed, head-to-head benchtop study provides somesupport to the notion that all threecommercially-available hysteroscopicmorcellators appear to be non-inferiorto each other in terms of the risk ofintravasation as a result of the observedintrauterine pressure being greater thanthe set pressure selected. This is impor-tant for clinical practice because manygynecologists do not have a choice as towhich device they can use—theirchoice is dictated by contracts signedbetween their hospitals and the manu-facturers of the devices.

Acknowledgements

The authors thank Jasmine L. Han-

key for editorial assistance in preparingthis manuscript. The authors would liketo thank Elite Research, LLC for inde-pendently validating our statisticalanalysis and providing expert statisticalconsultation.

Authors’ Disclosures

Both Dr. Howard and Dr. Stockwellhad full, complete, and unfetteredaccess to the raw data. No employee ofBoston Scientific Corporation partici-pated in the writing of the manuscript.Neither author received any compensa-tion from Boston Scientific Corpora-tion.

References

1. Marlow JL. Media and delivery systems.Obstet Gynecol Clin North Am 1995;22(3):409–22.2. Shveiky D, Rojansky N, Revel A, et al.Complications of hysteroscopic surgery:“Beyond the learning curve.” J Minim InvasiveGynecol 2007;14(2):218–22.3. Jansen FW, Vredevoogd CB, van Ulzen K,et al. Complications of hysteroscopy: Aprospective, multicenter study. ObstetGynecol 2000;96(2):266–70.4. Propst AM, Liberman RF, Harlow BL, etal. Complications of hysteroscopic surgery:Predicting patients at risk. Obstet Gynecol2000;96(4):517–20.5. Technology assessment No 7: Hys-teroscopy. Obstet Gynecol 2011;117(6):1486–91.6. Bradley LD. Complications in hys-teroscopy: Prevention, treatment and legalrisk. Curr Opin Obstet Gynecol 2002;14(4):409–15.7. Karaman E, Kolusari A, Cetin O, et al.What should the optimal intrauterine pressurebe during outpatient diagnostic hysteroscopy?A randomized comparative study. J ObstetGynecol Res 2017;43(5):902–8.Figure 2c.TruClear™ hysteroscope.

#1048 Howard FINAL


REFERENCES

STI

CONCLUSION

ACKNOWLEDGMENTS

AUTHORS’ DISCLOSURES

DISCUSSION

Documents

Intrauterine Pressure During Hysteroscopic Morcellation: A ... · An endoscopic optical lens is inserted through the cervix into the endometrial cavity to directly visu - alize and