DEPARTMENT OF HEALTH AND HUMAN SERVICES

	+ + + + +

	PUBLIC HEALTH SERVICE

	+ + + + +

	FOOD & DRUG ADMINISTRATION

	+ + + + +

	MEDICAL DEVICES ADVISORY COMMITTEE

	+ + + + +

	ORTHOPEDICS AND REHABILITATION DEVICES
	ADVISORY PANEL MEETING

	+ + + + +

	Thursday,
	November 4, 1999


		The meeting was held in Room 020B of the Center For Devices and Radiological Health, 9200 Corporate Boulevard, Rockville, Maryland 20850, Dr. Michael J. Yaszemski, Acting Panel Chair, presiding.


PRESENT:

MICHAEL J. YASZEMSKI, M.D., Ph.D., Acting Panel Chair
ALBERT ABOULAFIA, M.D., Voting Member
MARCUS P. BESSER, Ph.D., Approved Consultant
EDWARD Y. CHENG, M.D., Voting Member
BLAKE HANNAFORD, Ph.D., Approved Consultant
KINLEY LARNTZ, Ph.D., Approved Consultant
CATO T. LAURENCIN, M.D., Ph.D., Voting Member
RAYMOND SILKAITIS, Ph.D., Industry Representative
HARRY B. SKINNER, M.D., Ph.D., Voting Member
CEDRIC WALKER, Ph.D., P.E., Approved Consultant

FDA PARTICIPANTS:

HANY DEMIAN, M.S., Panel Executive Secretary
JIM DILLARD, M.S.
JOHN GOODE, M.S.
THOMAS P. GROSS, M.D. M.P.H.
MARK MELKERSON, M.S.
NEIL OGDEN, M.S.

	I-N-D-E-X


Call to Order	4

Opening Remarks	11

FDA Update From Last Meeting	11

Postmarket Evaluations at CDRH	13

Open Public Session	22

Reclassification of Constrained Total	23
Hip Arthroplasty Devices

	Petitioners Presentation	29
	FDA Presentation	49

Panel Discussion:

Deliberations on Reclassification of	71
Constrained Total Hip Arthroplasty Devices

Open Public Session	144

Computer Controlled Surgical Systems	150

	FDA Presentation and Questions	151
	Panel Discussion	158



	P-R-O-C-E-E-D-I-N-G-S
	(9:06 a.m.)
		MR. DEMIAN:  Good morning, everyone.  We're ready to begin the meeting of the Orthopedic and Rehabilitation Device Advisory Panel.
		My name is Hany Demian and I'm the Executive Secretary of this panel, and I'm a Senior Scientific Reviewer in the Orthopedics Devices Branch.
		I would like to remind everybody that you're requested to sign in on the attendance sheets at the tables by the door.  You may also pick up an agenda and information about today's meeting, including how to find out about future meeting dates and how to obtain meeting minutes or transcripts.
		I will now read two statements that are required to be read into the record, the Deputization of Temporary Voting Members Statement and the Conflict of Interest Statement.
		Pursuant to the authority granted under the Medical Devices Advisory Committee charter dated October 27, 1990, as amended April 20, '95, I appoint the following members for the Orthopedic and Rehabilitation Devices Panel for this meeting on November 4, '99:  Marcus Besser, Blake Hannaford, Kinley Larntz and Cedric Walker.
		For the record, these people are special Government employees and are consultants to the panel under the Medical Device Advisory Committee.  They have undergone the customary conflict of interest review and have reviewed the material to be considered at this meeting.  And this is signed David Feigal.
		Conflict of interest statement.  The following announcement addresses conflict of interest issues associated with this meeting and is made part of the record to preclude even the appearance of any impropriety.
		To determine if any conflicts existed, the Agency reviewed the submitted agenda for this meeting and all financial interests reported by committee participants.  The conflict of interest statutes prohibit special Government employees from participating in matters that could affect their or their employers' financial interest.
		However, the Agency has determined that participation of certain members and consultants, the need for whose services outweigh the potential conflict of interest involved, is in the best interest of the Government.
		Therefore, waivers have been granted for Drs. Harry Skinner, Cato Laurencin, Kinley Larntz for their interest in firms that could potentially be affected by the panel's recommendation.  Copies of these waivers may be obtained from the Agency's Freedom of Information Office, Room 12A-15 of the Parklawn Building.  
		We would like to note for the record that the Agency took into consideration other matters regarding Drs. Edward Cheng and Michael Yaszemski.  Each of these panelists reported financial interest in firms at issue, but in matters that are unrelated to  today's agenda.
		The Agency has determined, therefore, that they may participate fully in all discussions.
		In the event that the discussion involves any other product or firms not already on the agenda for which FDA's participant has a financial interest, the participant should excuse him or herself from such involvement and the exclusion will be noted for the record.
		With respect to all other participants, we ask, in the fairness, that all persons making statements and presentations disclose any current or previous financial involvement with any firms whose products they may wish to comment upon.
		Before turning this meeting over to Dr. Michael Yaszemski, our Acting Panel Chair, I would like to introduce our distinguished panel members who are generously giving their time to help FDA in matters being discussed today and other FDA staff seated at the table.
		We'll just go around the room and everybody introduce themselves and give where they're from and their interests.  Michael?
		DR. YASZEMSKI:  I'll start.  My name is Michael Yaszemski.  I'm an orthopedic surgeon and a chemical engineer.  I work at the Mayo Clinic.  My interests are adult reconstruction orthopedics, spine surgery on the clinical side, and bioresorbable polymers for use in bone regeneration on the research side.
		DR. LAURENCIN:  I'm Dr. Cato Laurencin.  I'm a Professor of Chemical Engineering at Drexel University and Professor of Orthopedic Surgery at MCP Hahnemann Medical School.  Interests are in drug delivery, polymeric materials, tissue engineering.
		DR. LARNTZ:  Kinley Larntz, Professor Emeritus, University of Minnesota.  I'm an Applied Statistician and my interests are experimental design clinical trials.
		DR. SKINNER:  Harry Skinner.  I'm Professor and Chair of Orthopedic Surgery at UC-Irvine and Professor of Mechanical and Aerospace Engineering at the College of Engineering at UCI.  And my research interests are gait analysis, adult joint reconstruction and finite element analysis.
		MR. DILLARD:  Jim Dillard.  I'm the Acting Division Director of the Division of General and Restorative Devices here in FDA, and my interests are all of the devices that come before this advisory committee.
		DR. SILKAITIS:  My name is Raymond Silkaitis.  I'm the Vice President of Regulatory Affairs for Gliatech.  I'm the industry representative.  I've been in the medical device industry for 20 years in the capacity of product development, research, clinical research and regulatory affairs.
		DR. WALKER:  I'm Cedric Walker.  I'm Professor of Biomedical Engineering at Tulane University and Chairman of Engineering Science.  My research interests are in the area of implantable medical devices, particularly electronic stimulation devices.
		DR. ABOULAFIA:  My name's Albert Aboulafia.  I'm an orthopedic surgeon with an area of interest in tumor surgery and orthopedic oncology.  I recently left Emory in Atlanta and am now at Sinai Hospital in Baltimore with the University of Maryland.
		DR. HANNAFORD:  My name is Blake Hannaford.  I'm a Professor of Electrical Engineering at the University of Washington in Seattle, and also Adjunct Professor of Bioengineering and Adjunct Professor of Surgery.  This is the only place where all those titles are relevant.
		And I do research on human interaction with robots and surgical biomechanics.
		DR. CHENG:  My name is Edward Cheng.  I'm an orthopedic surgeon at the University of Minnesota, and my interests are in musculoskeletal oncology and adult reconstructive orthopedics.
		DR. BESSER:  I'm Mark Besser.  I'm at Thomas Jefferson University in Department of Physical Therapy.  I'm a mechanical engineer.  Interests are in gait analysis and biomechanics.
		MR. DEMIAN:  Thank you.  We have one housekeeping order.  I would like to inform the panel that we have a new consumer representative, Ms. Cheryl Gartley.  She's the president of the Simon Foundation For Continence in Ohio.  However, at the last minute, due to a medical condition, she was unable to attend this meeting and, because of such short notice, we were unable to find a replacement.
		So, at this time, I would like to turn the meeting over to our chairman, Dr. Michael Yaszemski.
		DR. YASZEMSKI:  Thank you, Hany.
		Good morning.  My name is Michael Yaszemski.  I'll be the Acting Chairman for this meeting.
		I'd like to note for the record that the voting members present constitute a quorum, as required by 21 CFR Part 14.
		First, Mr. Mark Melkerson, Branch Chief of the Orthopedics Devices Branch, will provide us with an update from the last panel meeting.
		MR. MELKERSON:  Good morning.  This is Mark Melkerson, Branch Chief, Orthopedic Devices.  Actually, I'll be updating from the last couple of panel meetings.
		DR. YASZEMSKI:  Thank you.
		MR. MELKERSON:  As far as actions regarding previous devices that have come before the panel, the Norian SRS cement was approved December 28, 1998.  DePuy AcroMeds Lumbar I/F cage with VSP spinal system was approved in February of '99.
		The Sofamor Danek, Interfix threaded fusion device was also approved May 14, 1999.  And another reclassification petition was actually signed October 14th which reclassifies polymethyl methacrylate bone cement from Class II -- or into Class II from Class III transitional.
		As far as division staffing, already noted.  Jim Dillard is acting as our division director due to a death in another division.  Dr. Celia Witten will be acting as Division Director for Cardiovascular and Respiratory Devices.  During the interim, Jim Dillard will be our acting Division Director.
		As far as branch staffing in orthopedics, Ms. Jodi Nashman Anderson will be the team leader for bone and joint replacements and miscellaneous devices.  And for osteosynthesis and spinal implants, Ms. Erin Keith.
		That ends my update.  And if there's any questions, I'm available.
		Excuse me, one last -- recent 510(k) clearances may be of interest to the panel.  We have cleared a metal-on-metal semi-constraint hip prosthesis.  The manufacturer is Sulcer.  And we've also cleared a vertebral body replacement device, and that is a DePuy AcroMed product, a stackable cage with supplemental fixation.
		That ends my presentation.
		DR. YASZEMSKI:  Thank you, Mr. Melkerson.
		Seeing no questions, we'll next ask Dr. Thomas Gross of the Center For Devices and Radiologic Health to provide the panel with a presentation regarding post market evaluations.
		Dr. Gross.
		DR. GROSS:  Good morning.  My name is Tom Gross and I'm the director of the Division of Post Market Surveillance here at CDRH, and I'd like to take a few minutes of your time today to talk about post market evaluation.
		We at the center think it's important that the advisory panels are aware of post market programs and activities since they may directly affect your deliberations about a product's safety and effectiveness.
		Now, there are three key objectives for this presentation.  One, to describe a few of the key methods of post market evaluation.  Two, to present challenges in accomplishing post market evaluation.  And three, to describe the pivotal role that you play in this arena.
		I'm not sure why that's cut off, but, in any event, this title -- this slide entitled "From Design to Obsolescence" depicts three key points.  The first point is that it depicts the natural history of a medical device from design to lab bench testing, clinical testing, FDA review, and, importantly, post market evaluation.
		Secondly, it presents continual feedback loops throughout this process that leads to continual product improvement, and we think that post market evaluation has an important part to play in this process.
		The remainder of this talk will focus on three key programs in post market evaluation:  the MDR Program, post market surveillance under 522, and post approval studies under our PMA authority.  And the third point that this slide makes is that the clinical community and, importantly, yourselves play a key role in this process of continual product improvement.
		Now, as you all know, as products are released into the marketplace, questions of potential public health interest may arise.  They may be related to a product's long term safety, about a performance of the device in community practice as it moves outside the narrow confines of clinical trials into general community use.
		There may be concerns about effects of changes in user setting, such as moving from professional to home use; concerns about effects of changes in technology; and also concerns about adverse events or patterns of adverse events.
		Now let's talk about some of these programs that may address some of those issues, starting with the Medical Device Reporting Program, or MDR.  Now MDR is a nationwide passive surveillance system of voluntary and mandatory reporting.
		Voluntary reporting started in 1973, mandatory reporting in 1984.  And currently manufacturers must report deaths and serious injuries if a medical device may have caused or contributed to the event, and they're also required to report malfunctions.
		User facilities, and most notably hospitals and nursing homes, have to report deaths to the FDA and deaths and serious injuries to manufacturers.  Now beginning in the early '90s, FDA received about 100,000 medical device adverse event reports per year.
		And currently, all told in our database, we have slightly more than one million reports.  Now these reports are submitted on standardized forms and the information includes device specifics, event descriptions, pertinent dates and patient characteristics.
		Unfortunately, many of these reports often have very limited information.  Even basic information such as age and gender is missing from a large portion of reports.  Nonetheless, they can provide us critical signals, signals for which we'll take action.
		What are some of those actions?  The MDRs may lead us to directed inspections of manufacturers or facilities, product injunctions or seizures, product recalls (as in the case of surgical instruments being mislabeled), patient or physician notifications (as in the case of steam resterilization of zirconia ceramic femoral heads).  Also, it may lead to additional post market studies.  
		Now we have two authorities to base our requirements for a post market study.  One is Section 522 of FDAMA entitled "Post Market Surveillance," and one is under our PMA authority for post approval studies, better known as condition of approval studies.
		Now under Section 522, that was originally mandated in SMDA 1990, and it was changed significantly in FDAMA '97.  In the '90 version, there were actually categories and lists of devices, the manufacturers of which were required to do post market surveillance studies on, regardless of whether there were pertinent public health questions.
		Those categories and lists no longer exist in the FDAMA version, but we still have the authority, the discretionary authority, to order companies to perform post market studies if there are pertinent public health questions.
		Now, post approval studies or condition of approval studies refer strictly to PMA products.  Our 510(k) authority extends our coverage to Class II or III 510(k) products whose failure may present a public health problem.  Now, we see both authorities as a complement to our premarket efforts.  
		Now, in implementing the FDAMA version of post market surveillance, we publish criteria to help guide our considerations on when to impose post market surveillance on Class II or III products.  The principal criterion was that there had to be a critical public health question.
		Now, that could arise from "for-cause" issues such as adverse events, concerns about newer expanded conditions of use, concerns about effects of the evolution of the technology.  We also had to consider whether there were other, more pertinent post market strategies that could address -- better address the public health question of interest such as inspections or some aspect of the quality systems reg.
		Thirdly, we need to address whether the studies are practical and feasible.  Can we recruit the number of patients that we'd like to?  Can we recruit physicians to do the studies?  How will the data be used?
		This is particularly important for rapidly evolving technologies.  By the time we get the data, the data may be obsolete.  And what's the priority for this study?  We live in an era of limited resources.  We have to assess what the priority is, given the magnitude of the risk and benefit.
		Now once we decide to impose post market surveillance, there are a variety of study approaches -- study design approaches that may be chosen.  Obviously, the study design that is chosen should match the public health question of interest, and it should be the least burdensome approach.
		I've detailed a variety of approaches starting from the most general to the most sophisticated, most general being a detailed review of complaint history literature, non-clinical testing device, and so on and so forth.
		And occasionally, we may lead to randomized trials to address the public health question of interest.
		Now, we've experienced several frustrations in the post market period in instituting these studies, especially early in the life of 522, and we still experience some of these difficulties today.  Those being that the rapid evolution of technologies may make studies obsolete.  By the time we get the data, the data are obsolete.  
		There may be lack of incentives for industry to do these studies.  Industry may view these studies as being bearers of bad news.  There's nothing positive for industry in terms of doing these studies.  And we have to change that paradigm.
		There may be lack of interest in the clinical community in doing these studies, especially on mature technologies.  And early in the program, there were instances of a lack of clearly specified public health questions, especially for mature technologies that were required to be studied under the mandate, SMDA '90.
		Now what is the challenge to the advisory panel?  And really, the challenges to us all.  When considering post market studies, whether they're post approval or 522, we need to ensure that they're of primary importance; that they're practical, feasible; that the resources are warranted to do these studies.
		Obviously, we need to clearly specify the public health question.  And we need to note the clinical and regulatory relevance of answering the question:  What will we do with the data once received?  Are the data there to assure us that what we see in the post market arena are similar to what we know from premarket data?  Are they there to address residual questions?  And so on and so forth.
		This last slide depicts the future of MDR and post market surveillance.  With regard to medical device reporting, we're moving away from individual reporting of well-characterized events to summary reporting.  We're moving away from universal reporting by hospitals to focusing on a representative set of hospitals and sentinel reporting so that we can increase the quality of reporting.
		We're moving towards electronic interchange of reports, integration of our efforts with the quality systems regulation, and exchange of reports internationally.
		On the post market surveillance side, I've mentioned that we're applying a wider variety of study design approaches.  We'd like to work collaboratively with industry and the clinical community to achieve these studies.  And we're also attempting to get expanded access to relevant data sources to address these issues such as registries.
		That concludes my talk, and any questions?
		DR. YASZEMSKI:  Seeing no questions, thank you, Dr. Gross, for your presentation. 
		We'll now proceed with the open public hearing session of this meeting.  I'd like to ask at this time that all persons addressing the panel come forward and speak clearly into the microphone, as our transcriptionist is dependent upon this means for providing an accurate recording of this meeting.
		We're requesting that all persons making statements during the open public hearing session of the meeting disclose whether they have financial interests in any medical device company before making your presentation to the panel, in addition to stating your name and affiliation.
		Please state the nature of your financial interest, if any.  
		Is there anyone at this time wishing to address the panel?
		Since there are no requests to speak in the open public hearing, we will now proceed directly to the open committee discussion.  I would like to ask Mr. Jim Dillard, acting director of the Division of General and Restorative Devices, to provide an introduction to the concept of reclassification.
		MR. DILLARD:  Thank you, Dr. Yaszemski.
		I'd like to, with your permission, Dr. Yaszemski, do something before I actually introduce reclassification.  As Mark touched on a little bit, we had a colleague that was very dear to our heart and actually very dear to this panel, Dr. Tom Callahan, who passed away a couple weeks ago.
		And Dr. Callahan -- he started his career at the FDA in 1978 and immediately joined in the orthopedic area.  His background was in biomaterials.  He came as a researcher from institutions such as Stanford and Yale and had done quite a bit of work in biomaterials and biomaterials development.
		And when we got him at the FDA, it was a very good thing for us because he brought a lot of expertise.  He joined the orthopedics group in the late '70s and moved to become the Branch Chief for the Orthopedics Devices Branch, as well as the Associate Director in the Division of Surgical and Rehabilitation Devices, which, at the time, was the name of the division that housed the orthopedic and restorative group.
		And Dr. Callahan was the associate director over those two areas for quite a bit of time through the 1980s, as well as very instrumental in the regulatory effort for silicone gel-filled breast implants.  Tom headed up the review team.  
		And he, in the early '90s, moved on to become the Director of the Division of Cardiovascular Respiratory and Neurological Devices where he spent the remainder of his career.  And I just wanted to make mention for the record that we'll certainly miss Tom, and I know he was very important to this panel and very important to the FDA.
		And I thought it was worthwhile just noting some of the accomplishments in his career before this panel, so --
		DR. YASZEMSKI:  Thank you, Mr. Dillard, for those comments.
		MR. DILLARD:  Thank you.
		In terms of reclassification, this panel, I think, is well educated in reclassification.  We have spent a number of the past couple to three panels actually asking for your recommendation on reclassification petitions, and the reclassification petition today is no different.
		And what I wanted to do is just set a little bit of the ground rules again for everybody that what we're asking you to look at today is the reclassification of a product that was a preamendments Class III device that we called for PMA under our Section 515(b) parts of our regulations and called for PMAs -- and you will hear a little bit more of this in the presentations -- in the 1990s.
		What this reclassification petition is asking you all to do is to give a recommendation to the FDA as to whether or not there will be adequate controls that do not include premarket approval in order to go to the market -- or product development protocol -- in order to go to the market under the authorities that either include general controls, which are our Class I types of controls.
		Are they adequate alone to ensure the safety and effectiveness of these particular types of products?  Or could Class II controls, in addition to the Class I types of controls, help control for the risks and adequately ensure the safety and effectiveness of this product category in general?
		So what we'll be asking you to do -- and as usual, we will have Marjie Shulman, who is our reclassification coordinator for the Office of Device Evaluation, she will help lead you through the reclassification questionnaire after you have deliberated over the issues.
		Another bit of housekeeping is that everybody does need to fill out a reclassification questionnaire, although our Chair, Dr. Yaszemski, will have the official sheet, which he will try to develop consensus amongst you in order to give the FDA a recommendation for reclassification.
		But we do ask that all of you fill out a questionnaire so that, if there are any additional comments, we can consider them.  And you may not have them yet, but we will hand them out to you right before we actually go through the formal recommendation process from you as our panel.
		So with that, I think, Dr. Yaszemski, I will turn it back over to you.  I will be available, as well as Ms. Shulman, during the process for any help that you may need for the process.
		DR. YASZEMSKI:  Thank you very much, Mr. Dillard.
		We'll now begin the discussion of the reclassification petition for constrained hip arthroplasty devices.  We'll begin with the petitioner's presentation followed by the FDA presentation.
		This then will be followed by two lead panel member reviews.  Dr. Besser will discuss the preclinical aspects, and Dr. Skinner the clinical aspects.  
		After Drs. Besser and Skinner, we'll have a general panel discussion about this topic, followed by a panel discussion aimed at answering FDA's questions while going through the reclassification worksheet that Mr. Dillard just spoke to us about and the supplemental worksheet.
		We'll finish after the discussion by voting upon our recommendation. 
		I'd like to remind public observers at this meeting that, while this portion of the meeting is open to public observation, public attendees may not participate except at the specific request of the panel.
		We'll begin now with the petitioner presentation from the Orthopedic Surgical  Manufacturers Association with Mr. Lonnie Witham.  
		Mr. Witham.
		MR. WITHAM:  Thank you.
		Good morning.  As he said, my name is Lonnie Witham.  And this petition was submitted by the Orthopedic Surgical Manufacturers Association, also referred to as OSMA.
		I'm the immediate past president and currently a member of the board of directors of OSMA.  OSMA is a trade association comprised of 23 manufacturers of orthopedic implants, surgical instruments, and biological materials used in orthopedics.
		And this is the sixth orthopedic device reclassification petition OSMA has presented to the panel over the past three years.
		I'll give a brief overview of the petition, but I won't cover those areas to be addressed later by John Goode in the Agency's presentation.  I'll be followed by Dr. Thomas Brown from the University of Iowa, who will discuss the design considerations and recommended non-clinical testing.
		And then Dr. Andrew Brooker will discuss the clinical aspects of the petition.  The device description I'm going to skip over because that will be covered by John Goode later, as will the indications for use.
		He will cover more thoroughly the device identification, which is the current CFR classification of the metal/polymer constrained hip prosthesis, which is currently Class III, and the requested reclassification, which is Class II.
		With that, I'll skip to the summary of the reasons for reclassification.  Number one, the materials, geometry, articulating interface and fixation surfaces of the constrained hip prostheses are typically very similar to Class II semi-constrained prostheses.
		Number two, the risks to health have been identified and can be controlled by preclinical testing, labeling, guidance documents, published standards, and GMP and quality systems requirements.  These special and general controls are the same as those used to control Class II metal/polymer hip replacement prostheses.
		Number three, although the published experience with constrained hips is relatively small in comparison to that of the total hip arthroplasty in general, it's to be expected given the relatively limited patient population and indications for which the device is intended.
		The published results have been critically analyzed through a peer review process.  The results show consistency in pain relief, restoration of function, and reduction in recurrence of dislocation.
		The body of literature, along with the experience with this device in the commercial marketplace, demonstrates a safe and effective use when regulated with Class II controls.  And number four, since these constrained acetabular liners are not interchangeable from manufacturer to manufacturer, it's important to the patient and, thus, to the public health that a constrained hip liner can be supplied by each manufacturer of a total hip prosthesis.
		If a compatible constraint liner is not available to the surgeon, an entire well-fixed acetabular prosthesis may have to be removed from the patient in order to implant a constrained hip prosthesis.
		With that, we'll go the risks identified by the previous panel.  As you know, these devices were discussed in two previous PMAs, and that gave us a lot of groundwork for the risks to health and the special controls needed to minimize those risks.
		Certain adverse events and complications were noted in the literature and in the previous PMAs, and those were grouped into three major categories.  One is loss or reduction of joint function.  That includes loosening, revision of components, implant failure, fracture and wear.
		And to control those minimized risks, we have ASTM material standards, ASTM test methods, and we have three FDA guidance documents.  Another major category was adverse tissue reaction, which included osteosynthesis, sensitivity to metal implants.
		Again, we have ASTM material standards and FDA guidance documents.
		The third category was infection.  And special controls to minimize that risk was 510(k) sterility review guidance from the FDA.  There were additional risks identified: nerve impingement or damage, pain, vascular disorders, pulmonary embolism, gastrointestinal and genito-urinary complications.
		And these additional risks identified are associated with orthopedic surgery in general and are not necessarily unique to a constrained hip surgery.
		To summarize the list of special controls available, the following list of special controls available to minimize the risks to health identified by the petitioner and confirmed by a previous panel.  These special controls are in addition to the general controls applicable to all orthopedic implants.
		These special controls include ten ASTM standards for materials and test methods and six FDA guidance documents.  In addition, the FDA may require certain mechanical testing as part of the 510(k) premarket notification.  These tests will be described later by Dr. Brown.  
		The ASTM standards define implant material specifications and testing methods applicable to the constrained hip prosthesis.  Adherence to these standards and comparison of the results from these standard tests can control the risk to health of adverse tissue reaction, pain and/or loss of function, and revision by having the manufacturer use surgical implant quality materials and assuring that the device has acceptable performance through mechanical testing.
		FDA guidance documents provide guidance on how to meet general orthopedic device premarket notification or 510(k) requirements, including biocompatibility testing, sterility testing, mechanical testing, and physician and patient labeling.
		Use of the preclinical section of the FDA guidance documents can control the risk to health of adverse tissue reaction, infection, pain and/or loss of function, and revision by having manufacturers use surgical quality implant materials, adequately test and sterilize their devices, and provide adequate directions for use and patient information.
		And another control is -- another control is labeling.  The following indications for use, relative contraindications, warnings and precautions were identified by a previous panel for devices to be reclassified.
		We agree with the previous panel's labeling recommendations, and no new information is contained in this petition that would change the labeling previously established.  We also have the -- we'll skip to the relative contraindications.
		Three of those identified by a previous panel, which I'm sure many of you -- some of you participated in.  Warnings -- these are instructions to the surgeon on bending and contouring or modifying the device, improper alignment of the device, not to autoclave ultra-high molecular weight polyethylene, things of that nature which could cause an adverse reaction or early device failure.
		And the potential adverse effects, there are 12 of those identified by the previous panel.  We found no new information to change those and they'll stand as recorded by the previous panel.
		In conclusion, we believe the risk to health associated with the constrained hip can be adequately controlled with Class II regulatory requirements, and we hope the panel will agree and recommend these devices be reclassified.
		And with that, I would like to introduce Dr. Thomas Brown from the University of Iowa Hospital Biomechanics Laboratory, and he'll present the non-clinical testing.
		DR. BROWN:  Thanks very much.
		Yes, my name is Tom Brown.  I'm a mechanical engineer, Professor of Orthopedic Surgery in Biomedical Engineering at University of Iowa, and I direct their biomechanics lab.
		The OSMA folks asked me to come and present some of the preclinical testing data for these constrained devices.  Apparently they asked me to do this because our laboratory is active in the research involving total hip dislocations.
		I need -- I'm seeing these slides actually for the first time.  I need to point out that data are largely lacking in terms of the physiologic demands responsible for dislocations.  And most of you, I think, are aware of some of the factors that are pertinent here -- certainly issues of surgeon placement of the cups, issues of patient maneuvers and activities, and of course design issues on the device itself.
		Even though we don't have any hard numbers as to what the physiologic demands are, at least there are things that we can do in the laboratory to objectively measure intrinsic resistance to the dislocation and dissociation of these devices.
		And most of the -- back up a second, please.  Most of these laboratory tests are designed to evaluate what would be the failure modalities here, which I think are basically two that are specific to this issue.
		One is redislocation of the device.  And secondly, dissociation between the liner and the backing.  Responsible for the dissociation and dislocation are really pull-out sort of things and lever-out sort of things.
		So the testing that's designed for this designated as a lever-out resistance, a push-out resistance of the liner relative to the backing.  And then number (c) and (d) here, push-in and pull-out of the head and then the moment necessary to lever-out or toggle-out the femoral head.
		Okay, this lever-out resistance between the liner and the shell -- here basically the head and the acetabular shell are tested and basically a lever-out moment is applied in an MTS or Ingstrom-type testing.
		Slide, please.
		This test (b) basically is an axial push of the liner out of the shell backing.  This would be a test essentially of the locking mechanism.
		Slide, please.
		And then these (c) tests here are pertinent to the dislocation issue or relocation issue, an axial push-in and, more importantly, a pull-out of the shell relative to the head.  And this pull-out would obviously be a concern for these constrained devices, whereas it would not for a garden variety, semi-constrained device.
		Slide, please.
		And then lastly, a sort of a dislocation resistance.  This is a test performed to determine the torque that's needed to toggle the femoral head out of the socket during an impingement event.
		Slide, please.
		Okay, there's a relatively limited amount of testing that's been done here, most of it by in- house work in the manufacturers and a little bit from the Mt. Sinai lab, Seth Greenwald's.  Biomet's data for their device -- they have a hard number for the lever-out force necessary to disassociate.
		They have a hard number for the push-out resistance of the liner from the shell, as well as to push the head into the shell.  These obviously are going to be design specific.  As a point of reference -- let's see.  No, I think I want the next slide.
		There are three liners for which there is fair amount of data available.  Again, these are from the different manufacturers.  Pull-out data for the heads out of the shells; and then, I think most importantly, for dislocation, these toggle-out data.
		And as a point of reference, at least in our laboratory we've worked a lot with semi-constrained device lever-outs and typical numbers to dislocate are on the order of 70 inch-pounds.  So these things are all substantially more resistant to that than the semi-constrained device.
		And I think that's the end of my slides, so I guess I'm here to answer any biomechanical questions that might come up.
		Thank you.
		MR. WITHAM:  I'd like to introduce Dr. Andrew Brooker.  He's an orthopedic surgeon specializing in total joints.  He was Professor of Orthopedics at Johns Hopkins for 19 years, and he's now in private practice and has been for the last four years in Amarillo, Texas.
		He's a member of the Hip Society and AOA.
		Dr. Brooker.
		DR. BROOKER:  Thank you.
		Not having presented to this group before, I did my notes on an envelope while traveling here.  I thought first, with apologies to the orthopedic surgeons in your group, it might be beneficial to just briefly explain what we're talking about.
		The two types that Tom referred to of constrained liners are fundamentally either a ball-in-ball design where the liner fits into an existing metal shell and there are two balls within that.  The trunnion of the femoral head goes in and motion takes place at two intervals.
		The second is a ring-lock device where the polyethylene device locks into the metal cup.  The femoral head goes into that and a ring-lock extends over the tabs, thereby holding the head in.  The major differences that you see in the lever-out force occur because in the ring-lock device the poly extends all the way around the femoral head and is held with a ring-lock.
		The interesting thing about these is that they are manufacturer specific.  So, when you are confronted with an individual who has an existing total hip in and becomes what we call a "recurrent dislocator," you either have to go with that manufacturer's specific version of a constrained liner or, in the instance of a very large number of my patients, having used a system that does not provide a constrained liner, I have to remove the existing bone ingrowth liner in order to go to a constrained device.
		The indications -- indications are largely either to treat recurrent dislocation or to prevent anticipated dislocation or recurrent dislocation.  The population at risk -- and let's have this one slide -- we were able to come up with a number of studies that give an average risk of dislocation following the semi-constrained or standard total hip replacement.
		This falls in a range of one to six, with an average of 3.3%.  This is not the population that we're talking about using constrained liners on.  The population to use constrained liners is that group that -- the 3.3% that then go on to suffer recurrent dislocation.
		I, myself, over the years, have told my patients that dislocate after a primary hip replacement that their chances of redislocating are about 20%, somewhere in the 15 to 20% range.  
		I am not familiar with a study that puts a hard value on that because there are so many variables involved in the reason for a patient dislocating -- not only position of the implant, but age, weight, muscle strength and all that. 
		So what we're talking about in the population is basically perhaps in the range of 15% of the 3.3 group that dislocate.  The important consideration for this concept is that it provides a very successful way of treating a very difficult problem.  
		Most of these are elderly individuals, frequently overweight.  The typical patient that becomes a recurrent dislocator is a little, old lady living by herself on the cusp of going into a nursing home who goes home from rehab, falls within the first month post op, dislocates, tears out all of the supporting soft tissue structures in the back of the hip, comes in, is relocated and then goes on to recurrent dislocation.
		The options for treating her are conservative management, which is bracing, abduction supports and this sort of thing, which are, in my experience, wholly unsuccessful in the elderly and overweight.
		Imagine wearing an abduction brace made of polypropylene from around your waist to at least your knee if you're 80 years of age and 180 pounds.  It just isn't something that works very well or is commensurate with their activity.
		Reoperating to a semi-constrained or a standard total hip has certainly been attempted.  And if you study these individuals and become convinced the reason they're dislocating is because your cup version or your femoral version is way off, then going in and revising those is a significant operation, particularly if they're cemented stems, and often fraught with, I feel, increased problems postoperatively because of the loss of supporting structures, muscle injury, etc.
		Very commonly you will study these individuals and not see an obvious malalignment of either the cup or the stem and be forced to reoperate on them trying to use what we call "dial a prayer," which is a high wall or elevated liner that you put in a certain position hopefully to prevent dislocation.
		One of the largest problems that has been created in my practice is individuals who recurrently dislocate who then surgeons reoperate on and put a longer and longer neck on, both lengthening and lateralizing the hip as they do.
		I have a number of patients that have presented with recurrent dislocations who are already over an inch to an inch and a half long on the affected side because of attempts to stabilize them.  This is much akin to the weight placed in the ear at age 12, and by age 20 your ear's down at your waist.
		You continue to lengthen, these soft tissues continue to expand, and there's no end to that problem.  The beauty of the constrained device in treating recurrent dislocation is it's the only device I know that can allow you to shorten this leg, bringing it back to a normal length, yet gaining stability.
		In that sense, I would point out that I think there are some other indications for the constrained device that relate to the risk of dislocation or recurrent dislocation, and that includes particularly patients who have either had large tumor reconstructive or trauma surgery which leaves them without abductors.
		If you have, say, a proximal femoral replacement where you have no abductor balance, you don't properly decelerate the hip when you walk and, after a period of time, the hip becomes very lax and you get into recurrent dislocation.
		In those individuals, constrained liners provide a good way of treating a difficult problem.  Further, I feel there are a number of people at risk for dislocation -- the Alzheimer's-type patients and the elderly, modestly demented folks who are still functional.
		I have one individual who is an Alzheimer's patient, lives at home with his wife, had hip replacements and dislocated on one side, had a constrained liner allowed him to continue his activity even though he was functionally unable to really follow hip precautions.
		Slides.
		They've prepared some slides here just to show you a little bit how small the numbers really are.  Again, the J&J S-Rom is the double ball system.  Excuse me, the J&J S-Rom and the Biomet Ringloc are the two -- the ring-lock concept.  The Osteonics Omnifit is the double ball.  
		Again, the range simply reflects the range of people having hip replacements.  Although, if you live in Texas, the mean range of 156 to 177 pounds is like saying Larry Allen weighs 200 pounds.
		Go ahead.
		Follow up -- again, this is a difficult number to really hang much on because most of these people come to you recurrently dislocated, and to establish what their scores were pre and post becomes a little murky because do you go all the way back to their pre total hip, or do you just consider how they were total hip?  Many of them really haven't recovered long enough to even really establish a pre score.  
		Number of the cases that were revision cases, most all of these have been revision cases.  I would point out that the number being somewhat lower in the ring-lock.
		Over half of these are my cases.  And because the numbers become small and you start dealing with people who are either demented or Alzheimer's or who have had tumor surgery, that changes the statistics very rapidly.
		So fundamentally, they're pretty much all multiple dislocation of revisions.
		Recurrent dislocations in this group.  This is a series where the numbers are very low considering the population.  What you're doing is you're taking individuals that have already been recurrent dislocators, operating on them and putting a constrained device in them.
		Go ahead.
		In summary, all of the obvious things.  The redislocation rate is extremely low.  The other complications have been commensurate with revision surgery.  There's nothing particularly unique about this operation.
		In fact, this is an operation that, if you have a device that actually locks into the existing cup and you're satisfied with the existing cup and femoral version, it's an extremely straightforward operation and relatively short for the patient.
		In summary, I would simply say that I think that it would be good to have these available for all manufacturers' cups because there is a large population of otherwise healthy people who have ingrown acetabular cups who will require treatment of recurrent dislocation and it would be very advantageous to them to be able to do it without having to remove a solidly ingrown, porous-coated or bone ingrowth cup.
		Thank you.
		DR. YASZEMSKI:  Thank you very much.
		I'd like to ask Mr. Goode from the FDA to present the FDA's thoughts at this time.
		MR. GOODE:  Good morning.
		My name is John Goode and I'm with the Orthopedic Devices Branch and the lead reviewer for the metal/polymer constraint total hip prostheses reclassification petition.
		I'd like to thank the petitioner for their presentation.  
		Before I get started, I'd like to pass around what we've been referring to, which I have the Biomet and the Johnson & Johnson components, which are, as the petitioner described, the ring-lock variety where a ring locks the acetabular liner onto the femoral head.
		And then, I also have only one of these.  This is the Osteonics Omnifit device that is the bipolar type component.  I just have the acetabular part of this.  This would have a metal stem which came off this side and a metal shell which would lock it into the acetabulum.
		So I'll pass these around and you can take a better look at them.  So I have the Johnson & Johnson version on this side and I have the Biomet and the Osteonics versions on this side.
		I'll provide an overview of the premarket application history for metal/polymer constrained hips.  Then I will present the current reclassification for these devices and compare that with the petitioner's proposal.
		I'll identify the proposed indications for use and device description outlined in the petition.  I'll briefly summarize the supporting information.
		I'll give an update of the FDA's medical device reporting system and compare that with the risk to health identified in the original classification and in the petition, and list the types of special controls proposed by the petitioner to limit those risks.
		Finally, I'll present several specific questions the FDA has for the panel.
		Metal/polymer constrained total hip prostheses are preamendments, that is, devices available before the medical device amendments in 1976.  In July of 1982, after reviewing the recommendations of the orthopedic devices section of the surgical and rehabilitation devices panel, the FDA issued a proposed rule proposing to classify these devices as Class III.
		The final rule in which FDA classified these devices in Class III was published in September of 1987.  From September of 1987 to December of 1996, manufacturers were able to market these devices via 510(k) premarket notifications that the FDA determined to be substantially equivalent to legally marketed devices.
		During this period of time, FDA cleared five 510(k)s for these devices.  On September 7th of 1995, FDA published a proposed rule requiring the filing of a premarket approval, a PMA, or a notice of completion of a product development protocol, or a PDP, for these devices.
		The comment period for the proposed rule closed on January 5th of 1996.  And according to the FDA Dockets Management Branch, they received no new comments regarding constrained hip prostheses.  In September of 1996, FDA published the final rule for these devices requiring a PDP -- a completed PDP or a PMA by December 26, 1996.
		Two orthopedic companies, Johnson & Johnson and Osteonics, Inc., filed PMAs for their constrained hip prostheses and both PMAs were approved in June of 1997.  The current classification states that a hip joint metal/polymer constrained cemented or uncemented prosthesis is a device intended to be implanted to replace a hip joint.
		This device prevents dislocation in more than one anatomic plane and has components that are linked together.  This generic type of device includes prostheses that have a femoral component made of alloys such as cobalt-chromium-molybdenum, and a acetabular component made of ultra-high-molecular-weight polyethylene.
		This generic type of device is intended for use with or without bone cement.  This device is not intended for biological fixation.  That is the current classification and the sponsors -- or the petitioner's proposed classification is very similar to the current version.
		The petitioner's proposal provides a definition regarding which implants are to be included or excluded from this classification.  In the petitioner's proposed classification, ultra-high-molecular-weight polyethylene acetabular component may be used with or without a metal shell.
		That's one of the changes that's bolded on the screen.  In addition, they eliminated the statement that this device is not intended for biological fixation.  That was also included in the original classification definition.
		The petitioner has proposed a change in the classification of these devices from Class III to Class II.  Total hip prostheses or orthopedic reconstructive devices intended to replace the principal articulating surfaces of the hip joint, that is the femoral head and the acetabulum.
		Hip replacement is typically performed when the surfaces of the femoral head and acetabulum have been severely damaged by degenerative joint disease or traumatic injury.  The main objectives of this procedure are to relieve pain and restore function.
		Constrained hip devices are a subset of total hip replacement devices.  And while they are used to relieve pain and restore function, they are made for a patient that is at high risk to dislocate the femoral head from the acetabulum.
		Therefore, the specific indications for use proposed by the petitioner for the metal/polymer constrained hip devices are for patients at high risk of dislocation due to a history of prior dislocation, bone loss, soft tissue laxity, neuromuscular disease, or intra-operative instability.
		The proposed device description for constrained total hip prostheses include a metallic stemmed femoral component that is fixed in the femoral intermedullary canal with or without bone cement, an acetabular component that consists of a polyethylene constrained liner that may be used with our without a metal shell component, and fixation of the acetabular component in the acetabulum is achieved with or without bone cement.
		The femoral and acetabular components are linked together, typically by a locking ring that secures the polyethylene constrained liner around the femoral head (for example, the Biomet and the Johnson & Johnson device) or a bipolar component like the Osteonics device.
		This linkage stabilizes the hip joint and provides resistance to dislocation.  The constraining polyethylene liner retains the head of the femoral component.  This reduces the travel distance of the femoral neck, and therefore the range of motion of the hip joint is reduced as compared to a semi-constrained total hip.
		As the petitioner presented, there were nine clinical articles containing information on overall outcome and complications provided in the petition.  
		Of the nine articles, five included information on the Johnson & Johnson S-Rom device, two included information on the Osteonics Omnifit device, and two included information on other preamendment constrained devices, the SRN and the Russin-Sivash devices.
		The petitioner also provided unpublished information regarding the Biomet Ringloc constrained hip prostheses.  The petitioner provided a bibliography on the available literature, as well as copies of the articles in the petition that each panel member has received.
		The sponsor has stated that a literature search was performed using orthoguide.com that include a Medline search designed for orthopedics.  The keywords searched were "constrained" and "hip."  The search was for all articles from 1967 to the present.
		All articles that contain information on constrained hips were included in the petition, and there were a total of nine articles identified.  The second search that the sponsor performed used the key words "hip" and "dislocation."
		These references were to establish an overall dislocation rate following semi-constraint total hip arthroplasty.  Nine references were chosen because they reviewed a large number of cases and they were already presented by the petitioner.
		Now I will summarize the information gathered from FDA's medical device reporting system.  The MDR system can give us an indication of the types and relative incidence of various adverse events, but there are limitations to what the medical device reports can tell us.
		Some events go unreported either because the manufacturer doesn't find out about them, or it is determined that the event is unrelated to the device.  The reporting period summarized in the petition was from January 1985 through December 1998.
		There were a total of 91 medical device reports for constrained hip devices.  Fifty-six of the 91 reports concerned dislocations.  Twelve dislocations occurred during normal activities.  Ten occurred due to a dislocation of the femoral head from the polyethylene liner, and it was suspected that this was due to impingement of the femoral neck on the acetabular rim.
		Nine dislocations were due to misalignment of the components, and 25 were of unknown cause.  Eleven reports involved the disengagement of the polyethylene liner from the metal shell.  Seven reports were concerning a broken locking ring that holds the polyethylene liner onto the metal head.
		Five reports were concerning revision.  And there were other reports in the one to two category that were either ring migration, broken insert, cement loosening, tapers unlocked, liner wear, device split, poor liner fit, the ring wouldn't fit, and the size was mislabeled.
		In 1982, the following risks to health were identified in the Federal Register notice proposing constrained hip prostheses for Class III, and these were considered by the original panel.
		The petitioner has already presented these, the loss or reduction of joint function, adverse tissue reaction and infection.  In addition to those risks, the petitioner provided the following list of risks to health from the literature and those reported under MDR.
		And these include loosening or revision of components; dislocation; implant failure, fracture and wear; osteolysis; sensitivity to metal implants; infection; nerve impingement or damage; pain; vascular disorders; pulmonary embolism; gastrointestinal and genitourinary complications.
		The risks identified by the petitioner are more specifically delineated, but still appear to fall in the broader categories first identified by the original classification panel.  However, there may be additional risks of which you are aware, and the questions I will read later will include a request for you to identify any additional risks to health for constrained prostheses.
		In order to control the risks to health, the petitioner has identified various types of special controls to ensure the safety and effectiveness of constrained hip prostheses as Class II devices.  
		These include conformance to consensus standards such as the ASTM and isomaterial standards presented by the petitioner, FDA guidance documents, the preclinical component testing as discussed by Dr. Brooker, I believe -- or Dr. Tom Brown, and labeling to ensure the devices' proper use in appropriate patients.
		As you are considering the risk posed by these devices, you may identify other special controls you find to be appropriate.  You will be able to add to these -- add these to the list when you fill out the general device classification questionnaire.
		General controls such as good manufacturing practices and design controls may also be sufficient to limit some of these risks.  
		Now that I've provided some information on constrained hip prostheses, I would like to address five questions to the panel -- I'm sorry, four questions to the panel.
		Each of the members of the panel should have a copy of these questions in your packet of information.  Your answers to these questions will be recorded on the reclassification questionnaire after your preliminary deliberation.
		The first question is:  Does the petitioner's proposed classification sufficiently describe metal/polymer constrained hip devices?  If not, what other types of descriptive information should be included in the classification definition?
		And this will be question S-1 in your questionnaire.
		Number two:  Based on the known clinical information, for which patient population(s) should constrained hip devices be indicated for use?  This is question S-4 on your questionnaire.
		Number three:  Risks to health have been identified by the petitioner, the previous panel, and the medical device reports.  Have all of the risks to health for constrained hip prostheses been identified?  And if not, what additional risks should be described?
		And the final question:  The original classification included devices to be fixed with or without bone cement, but specifically excluded devices intended for biological fixation.  What impact does the means of fixation have on constrained devices (for example, cementing, hydroxyapatite coating, porous coating, or press-fit)?
		Has the petitioner provided sufficient information to reclassify devices intended for cemented, uncemented and/or biological fixation?
		I'd like to thank you for your attention.  I'll now turn the floor back over to the panel chair for discussion.
		DR. YASZEMSKI:  Thank you, Mr. Goode.
		This is Dr. Yaszemski speaking.  I'm going to ask at this point, since everybody on the schedule has been identified by their presentation, and since we are now going to transition into a general discussion in which it won't be apparent to the transcriptionist who is speaking, I will ask all the panel members and others who approach the microphone to identify themselves before they speak and ask your tolerance with me in advance if I -- I will try to introduce you if you neglect to do it for yourself.
		Let's all please try to identify ourselves for the transcriptionist, and I'll try to pick up on it if you don't.
		We're going to begin now with two reviews by our panel lead reviewers, first by Dr. Besser regarding the preclinical status, and then followed by Dr. Skinner with respect to clinical considerations.
		Dr. Besser.
		DR. BESSER:  Yes, this is Mark Besser.  This is Mark Besser.  The preclinical testing that's been described by both the petitioner and referred to by the FDA presenter, I'm not going to reread it again.  Essentially the five -- the testing falls into both materials testing and I think that the ASTM standards for the materials involved, the materials have been tested.
		And I think the preclinical testing for the materials doesn't need additional addressing by this panel.  For the actual devices, looking at both the shell liner disassociation, the push-out and lever-out tests that have been described, and there are standards for those.
		And for the femoral head dislocation, both the toggle-out force or torque and the push-in and pull-out forces have also been described by the standards.  I don't think that -- I had no further comments or requirements for additional preclinical testing to be done for these devices.
		I do think that for each design and each new design presented it's essential that preclinical testing be done for that design.  And you can't say that this is similar to previous designs and assume that you're going to end up with the same resistance to either liner disassociation or femoral head dislocation just because the design looks similar.
		So with the standards for preclinical testing that are currently in place, I think that no additional requirements for preclinical testing are necessary at this time.
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		We'll now ask Dr. Skinner to present -- Dr. Yaszemski here will ask Dr. Skinner to present his clinical discussion.
		DR. SKINNER:  For the record, my name is Harry Skinner again.  
		The nice thing about talking last is that you don't have to say anything.  I can just sit down now and say it's all been said. 
		Next slide.
		Just a little review.  The incidence of dislocation, according to Morrey, is about 3%, and it doesn't seem like there's a significant learning curve.  And it's a devastating problem for people who have a dislocation.
		It's a devastating problem for the surgeon, too.  There are many, many risk factors that have been identified to go along with dislocation, and those risk factors come down to -- the largely important ones are surgical approach, component position, component design -- and by that I mean things like sleeves, skirts, extended liners.
		And then the patient has a significant effect on this.  I think Dr. Brooker alluded to this.  Patient's ability to comply is probably one of the major problems I see in dislocations that I take care of.
		And the diagnosis is significant too.  Ehlers-Danlos, Parkinson's Disease, things like that are significant problems that lead to the need for things like constrained cups.
		Next, Mark.
		The constrained liner is an alternative when the component position is acceptable, when you can't use something like an extended liner, an extra thick liner, increased neck length, etc.  Many things that you can use to take care of dislocations.
		Next.
		Well, regarding the adequacy of information to support reclassification, we've heard about this to a great extent.  And with the huge total hip literature that's out there, which are already Class II devices, I think that the issues about polyethylene surfaces, metal surfaces, attachment of the liner to the inside of the -- to the plastic, all of these things have been pretty well taken care of with the total hip literature.
		And I think that what we have to deal with here is the issues relating to constraint, which is a small part of that whole topic.  And as has already been stated, there's a very small constrained liner of literature.
		Next.
		So the issues are the preclinical data, the appropriate patient population and special controls.  On the right you see a slide that shows one of the S-Rom components in place, and that ring is what you see on x-ray with one of these things.
		Next slide.
		With the preclinical data, I think the important things are the range of motion of the device and the arc of motion of device, which are really different things because these things, as you noticed when they were passed around, have offsets to them, so that the -- while the arc of motion may be 70 degrees -- the range of motion of the device may be 70 degrees, the arc of motion might be from 20 degrees of flexion to 90 degrees of flexion, or the reverse.
		So where the arc is, is quite important in these things.  And that's something that I think we have to address in the final product literature that's on this stuff.  The pull-out and lever-out things are important for the liner and the cup, the liner and the head, and the last thing that was addressed just recently was the cup and the bone.
		Next slide.
		The literature's already shown us the data for these.  Tom Brown mentioned these numbers.  And I think what these numbers show us -- and I think Dr. Besser alluded to these, too -- is that these two devices anyway, the Osteonics and the S-Rom, are pretty successful devices on the market at the present time, and I think they give us a general range for what is required for any new devices that come onto the market.  And I think that is what the goalpost should be that the new products that would presumably come along in a Class II situation would have to jump over.
		Next slide.
		The cup/bone is something that is a very serious problem with these things.  Generally the manufacturers -- I think that's Osteonics and Joint Medical Products, Johnson & Jonson -- have tried to make it so that there's been a problem -- if there's a problem either at the femoral head plastic or the plastic shell rather than at the cup/bone interface because that's better for the patient than having the cup come completely out of the pelvis with a large piece of bone or even perhaps even break the pelvis.
		So the issue of adequate fixation of the cup on the bone is a significant issue with these particular devices much more than with any other device. 
		Next slide.
		The head/neck geometry is significant.  Many of the heads that are available today have cutouts at the base where it goes into the Morse taper.  And that's going to change the lever-out or toggle-out effect of the head, and I think that has to be taken into account when these things are tested.
		Next slide.
		Indications for use, as suggested by the manufacturer, I think are very good, but I think there should be an additional caveat because, at the very minimum, the rep who is in the room with the surgeon who will know this information, because the surgeon certainly won't know it, will be able to say, "Doctor, is there any other way that you can stabilize the hip from dislocation?"
		Because that's what the package insert indication says, and that's why I've suggested the statement at the bottom.  And you can see the slide on the right what happens with one of those Johnson & Johnson cups when they dislocate.  That wasn't one of my cases, by the way.
		Next slide.
		Special controls -- I think they've been gone over quite adequately.  The ASTM standards, the special guidance documents for the FDA, but I think also there has to be an education and training process for the surgeon.  
		Because, for instance, for the J&J product, if you put the ring on backwards, it will almost certainly fail.  And it's a subtle difference when you put that metal locking ring on, and I'm sure the same thing happens with the Biomet process.
		So I think those are things that have to be addressed, too.
		Thank you very much.
		DR. YASZEMSKI:  Dr. Yaszemski here. 
		Thank you, Dr. Skinner.
		We're now going to proceed to the general panel discussion.  I'd like to take a moment and outline what that will consist of.  
		First we're going to go around the table and ask each panel member whether they have any general comments related to the presentations they've heard from either the petitioner, the FDA or our preclinical and clinical lead reviewers.
		During that time, I would ask the panel members to begin to consider the general data sheet and the supplemental data sheet, because our task will be to come to a consensus on how these sheets are filled out and to make a recommendation to the FDA as to what the filled out sheets should contain.
		So please, if you would, while we're going around the table the first time, start giving consideration to what you feel should be included in those sheets.  And I would ask each panel member to fill their names in at the top of the sheets because the FDA will be collecting these sheets from us at the conclusion of our deliberations.
		After we've done that, we'll ask the FDA to put up the specific questions posed to us, and we'll go around the table again to ask the panel what we feel should be our answers to those specific questions, and those answers will be put down in the form of the filled out sections of both the general and supplemental data sheets.
		I'd like to start now, if I could, by asking Dr. Cheng to begin the general discussion.  And then we'll go around the table in clockwise fashion and we'll go then to Dr. Hannaford and Dr. Aboulafia and the other panel members in order.
		Dr. Cheng, do you have any general comments on what we've heard this morning?
		DR. CHENG:  I have some general comments, but first I'd like to have your permission to ask a question of OSMA.
		DR. YASZEMSKI:  So granted.
		DR. CHENG:  I'm wondering, is there any data which you've presented, perhaps I've missed, that addresses cemented polyethylene cups, either metal or metal backed?  I believe these are all ingrowth, is that correct?
		Does anyone care to address that question?
		Let me give you my general comments and he can answer that in a moment.  I think it's a valuable tool in the surgeon's armamentarium to have these, but I think it's rarely ever required -- hopefully rarely ever required.
		And hopefully we don't have to use them very often because there is a high incidence of potential problems, as the MDRs indicated, and you're dealing with a very difficult population of patients.
		I would advise the FDA to consider changing the indications and adding a statement much as Harry said.  I wrote a different one that just said that these only be considered after all other possible issues, such as component malposition, placement, leg lengths and trochanteric function have been addressed.
		I would list a contraindication, which is specifically the malposition of a component.  I'd be concerned that surgeons would use this as a simple means to address the dislocation when perhaps there are other issues that should be addressed first.
		I think the preclinical testing is very important to be done because every -- the testing is device-specific and there are a number of cups where we know that the liner disassociation has a lower threshold for occurring than for other specific manufactured devices, and so the preclinical testing would be important.
		And I think the rationale that OSMA presented for reclassification is reasonable.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Do we have a comment from the petitioners regarding Dr. Cheng's question.
		Mr. Witham, thank you.
		MR. WITHAM:  We did have some information in the petition concerning the number of cemented and uncemented procedures from the published articles, but it was very difficult sometimes to determine what the fixation method was.
		For acetabular cups, we were able to, from the published articles, identify 207 uncemented and 15 cemented.  The femoral stems, there were 127 cemented and 18 uncemented, which is just about reversed, and 79 undetermined.
		The Biomet study, there were 60 uncemented, 22 cemented acetabular cups.  And since it was done retrospectively, for the most part, we had 72 that were undetermined.  So there's a mixture, but the ones we were able to identify are -- there are more uncemented than cemented on the acetabular site.
		DR. YASZEMSKI:  Thank you, Dr. Witham.
		Dr. Cheng.
		DR. CHENG:  The reason I asked that question is, the very first question the FDA asked us is if the proposed classification sufficiently describes it, and I'm not sure the cemented cups are appropriate for use for this.  Or at least there's minimal data that's been presented in that regard.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Dr. Hannaford.
		DR. HANNAFORD:  I think the issues here are primarily medical ones.  It appears to be a very mature technology.  But I would say that, from an engineering perspective, I would agree with the previous two panelists that this preclinical mechanical testing would certainly seem warranted.
		It would seem very necessary not to -- I guess what I mean is it would seem clear that each different design is going to have different mechanical properties and should -- which should be documented and carefully tested.
		DR. YASZEMSKI:  Thank you, Dr. Hannaford.
		Dr. Aboulafia.
		DR. ABOULAFIA:  Albert Aboulafia.  I have no general comments at this time, although I think we will discuss the modification of other methods likely to fail and things as time goes on, and I'll reserve that for later.
		DR. YASZEMSKI:  Thank you, Dr. Aboulafia.
		Dr. Walker.
		DR. WALKER:  Dr. Skinner's comments I think were excellent.  This is a small population.  It's in the patient's best interest to, in all cases, keep a functional acetabular cup rather than having to revise it.  
		I noticed that the original Class III guidelines forbade biological fixation and now biological fixation is being included, but I haven't heard any words today about testing or proof that the increased loading when there's -- when you run out to the end of range of motion would still allow biological fixation to take place.
		And I guess my question is more for the orthopedic surgeons.  Does it ever happen that you run out of range of motion with a constrained device? Clearly, with an unconstrained device, it's unlikely.  And is that going to affect the fixation of the femoral component if you do?
		Maybe one of the orthopedic surgeons could talk about that.
		DR. YASZEMSKI:  Would anybody like -- Yaszemski.  Would anybody like specifically to answer Dr. Walker's question at this time, or shall we include that in the specific discussion when we get to that panel question?
		Dr. Brooker.
		DR. BROOKER:  I'm sure if we go far enough around the panel, Dr. Skinner's going to comment on this because this is the direction he was heading with his slides.  But the only time that we have seen clinically where there has been a problem with range of motion was in the initial use of these with skirted implants.  In other words, the elongated femoral heads that effectively thickened the neck and therefore caused impingement to occur at a very much earlier time.
		And I think that is -- particularly in the ring lock devices, I think that's clearly a contraindication for using those devices, and I think that should be a consideration.
		I can't resist the temptation to address one issue.  I think that when you consider further verbiage about evaluating other methods, one of the things to consider, or way to put it, is you might want to state that the constrained device may be indicated if the femoral head and cup are in a satisfactory position.
		Because if the femoral head and cup aren't in a satisfactory position, then you're already behind the eight ball in terms of your range of motion.  On the other hand, there are those people in whom it is in position where you might not consider soft tissue or trochanteric osteotomy.
		So I think one of the centerpieces here -- and again, I'm stealing Dr. Skinner's thunder because I think he's probably heading in that direction -- is if the patient has good aversion and alignment of both the acetabular device and the femoral device in place and it's still dislocating.
		DR. YASZEMSKI:  Thank you, Dr. Brooker.
		Dr. Walker, does that answer your question?
		DR. WALKER:  Yes, maybe we can get to the biological fixation later.
		DR. YASZEMSKI:  Thank you.
		Dr. Silkaitis.
		Dr. Brooker, can we come back to that when we come around the room again?  I think we're going to expand on that as we go around, but thank you.
		Dr. Silkaitis.
		DR. SILKAITIS:  Yes, this is Ray Silkaitis.  And I guess my question is maybe directed more towards the FDA.  We're looking at this particular design, cemented, uncemented, various characteristics of the constrained hip, and we're making recommendations regarding its indications and risks associated with it.
		I guess my question is that there are already two devices approved by PMA; how does that fit into the evaluations that are being performed here?
		DR. YASZEMSKI:  Mr. Dillard.
		MR. DILLARD:  Yes, Jim Dillard.
		I believe, Dr. Silkaitis, in terms of a regulatory situation and perhaps a scientific situation -- and maybe let me break those out a little bit, although they're very much together.  
		The regulatory situation is if these products, in a general class of products, are reclassified to Class II, and the two existing PMA products would be converted to a Class II type of device, and that in the future, if any modifications were to be made to those devices, the submission of a 510(k) premarket notification would be necessary instead of a supplement to the premarket approval application.
		One of the other things that goes along with a reclassification is that the potential post market and quality system regulation components that go with a Class III device do change when you go from a Class III to a Class II or a Class I.
		Most of the post market requirements are increased in the Class III category, and they are somewhat diminished in the Class II.  Not from the standpoint of not still being considered -- you still must meet the quality system regulation and you still must meet design control activities as a Class II device, as you would as a Class III device.
		Reporting requirements are somewhat diminished under the PMA requirements.  You have to submit annual reports on your device.  That requirement is not there for Class II premarket notification products.
		From the standpoint of the science, the threshold for decision making, as everybody probably well knows, is a little bit different.  There is not an absolute requirement for determination of reasonable assurance of safety and effectiveness under a 510(k) premarket notification.  The standard is to demonstrate that you are substantially equivalent to a device that's on the market.
		So, in terms of the type of prospective data that may be required under a PMA to demonstrate reasonable assurance of safety and effectiveness, the target would more be on are you equivalent to, and the two devices that are approved are really the two devices that perhaps will define that category, are you equivalent to two devices that are legally on the market, and those two devices are currently legally on the market.
		So they will sort of set the benchmark.
		DR. YASZEMSKI:  That's Dr. Silkaitis talking again, excuse me.
		DR. SILKAITIS:  I'm sorry, thank you.
		That's with the guidance documents that are available in terms of the requirements for preclinical testing that all that is met under the 510(k), is that right?
		MR. DILLARD:  Jim Dillard.  
		The guidance documents don't necessarily set down requirements.  They are intended as guidance documents.  Although, if they are recognized as special controls, they will be formalized as something that the manufacturer needs to consider in a 510(k) premarket notification.
		So substantial equivalents -- or the idea of a 510(k) and substantial equivalents really sets down more the benchmark, I would say.  And those products that are on the market and those testing performance criteria that those products currently meet become more the standard, I think, in terms of substantial equivalence.
		DR. SILKAITIS:  So we're looking at decisions today affecting the entire class, even those that have been previously approved?
		MR. DILLARD:  Yes.
		DR. SILKAITIS:  Thank you.
		DR. YASZEMSKI:  Thanks, Dr. Silkaitis.  Anything further?
		DR. SILKAITIS:  No.
		DR. YASZEMSKI:  Mr. Dillard, have you any further comments?
		MR. DILLARD:  No, thank you.
		DR. YASZEMSKI:  Thank you.
		Dr. Skinner.
		DR. SKINNER:  Just a couple comments.  Dr. Skinner.
		First of all, I think the consensus of the orthopedic surgeons who have used these devices or who have at least experienced patients with dislocations is that these shouldn't be used routinely, and I think that's something that we would like to get across in the Class II description.
		When it comes to the preclinical data, what I've mentioned in the past discussion, I think that the preclinical data that's important from a biomechanical viewpoint is the toggle-out and lever-out type of failures of the devices.
		And I think that, for instance, the push-out of the cup is -- well, to use a push-out test pushing out through the hole in the back of the cup I think is kind of like pulling our leg.
		(Laughter.)
		It's not as valuable as you might think.  I think that toggle-out tests should also include heads with skirts and heads with chamfers on them because both of these may affect the toggle-out significantly.
		I think an important thing to document in the information that goes to the surgeon is the range of motion in these situations, too.  As Dr. Brooker alluded to, you put a skirt on one of these things and the range of motion, which might otherwise be 70 degrees, might drop down to 50 degrees, and that puts the patient at a very high risk of dislocation.
		Anybody that's done hip revision surgery knows that there are times when you'd be happy to be able to leave the operating room feeling comfortable that there's 50 degrees motion and it's not dislocating, but you would rather not have that happen and you'd rather try to keep -- you'd rather try to avoid that.  Knowing about that gives you a little bit more information to deal with while you're working on this problem.  
		Again, I'd like to come back to the cup/bone interface.  I think that it should be discouraged to use these inserts in cups that don't have a means of resisting tension between the cup and the bone.
		I think porous coating, particularly if it's ingrown, is certainly one way of accomplishing that.  I think cement would do that.  I think that screws would do that.  I'm not sure that a relatively smooth surface with hydroxyapatite on it would do that.
		So those are my comments.
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Dr. Larntz.
		DR. LARNTZ:  Kinley Larntz.
		And I'm a statistician, so I have to look at things from that point of view, or I choose to anyway.  I did read somewhere it says here, it says since -- this is from the manufacturer, or from the manufacturer association -- "Since the FDA classifies these devices into Class III, the development of devices and surgical technique has continued and a considerable body of published clinical results have appeared in the peer review literature."
		This body of "new information" provides the grounds for the present petition.  If that's what the present -- if that's the grounds, then that's not here because I don't consider this a considerable body of literature by any stretch, and I think you said that.
		There's almost been nothing done.  Very, very small number of cases providing the three -- for three devices, two of which are already approved.  What's interesting to me though is also, as a statistician -- and we all get reputations as statisticians for nitpicking.
		Well, I'll do it.  And in fact, in fact, there's double counting even in the materials presented.  One of the studies is actually a subset of the other study, and then they add up the cases.  That doesn't fly with me, and I think it's very clear if you read the articles, which we could not do for one of the devices, that, in fact, you'll see that double counting's there.  For whatever reason, there was some summaries going on today that said 90% of the cases were revisions.
		Again, if you go back and read the articles, eliminate the double counting, 25% of the cases are primary, 75% are revisions.  Now 25% is not 10%.  That's all I'm going to say.  And besides that, last point, if you look at the two approved devices, dislocation rates are very different -- very, very different.
		Now, what does that say?  Well, that says to me, as a statistician, well that says these devices are very different.  And so they have to be considered individually and carefully, and I'm not here to say how that individual and careful decision making should be done, but it's very clear that the two approved devices have very different abilities to function with respect to dislocation.
		Also, I said that was the last point.  I apologize.  None of the data for the two devices is any later than 1993.  So new information -- I don't know how new information counts in this day of electronic and wonderful things accumulating very fast.
		Much of the data that was published -- and even the data that was published in 1998, the series goes from 1988 to 1993.  That's the case where one series is a subset of the other.  So, enough said.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.
		Dr. Laurencin.
		DR. LAURENCIN:  Laurencin.
		My comments really just echo Dr. Skinner's and also Dr. Larntz in terms of the area.  Just want to underscore the fact, in terms of general comments, that the area is very, very important because, as we see, revision rates start to creep up over the years while the rates of -- the rates in which this is used now is low.
		As revision rates creep up and as we see more bone loss, surgeons will be more and more tempted to use these sorts of devices, and as more manufacturers make these devices and bring them into the operating room, I think that surgeons will be more tempted to use these sorts of devices.
		So I think it's going to be very important that really stringent criteria be given in terms of their use.  That should be written into -- you know, if there's a reclassification, written into that.
		Also, on the issue of biological fixation, it really does appear that initial fixation is even more important in these devices than in the semi-constrained types of devices, and so I think that it's going to be important that the committee consider really establishing criteria in terms of biological devices.	
		For instance, I was looking at a couple of the reviews and Cameron's review -- his recommendation is that any biologically fixed device has to be augmented with screws because of the early failures that he's seen with biological fixation devices and the feeling that, with the extra loads that are placed there, that screws should be used.
		So I think that should also be considered.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		We're next going to go around again with the specific intention of coming to a consensus on both the general device classification questionnaire and the supplemental data sheet.  
		I'd like to make one housekeeping announcement, if I can, prior to starting.  And that is, it appears that we perhaps will have a chance to move lunch up a little bit, so I'm going to ask in advance that we consider having lunch from 12:00 to 1:00 and starting the afternoon session at 1:00, and we'll adjust that as we go along, but we'll at least try to do that if we continue on a fairly rapid pace.
		Secondly, with respect to the two data sheets, what I would ask as we go around -- we're going to give each panel member an opportunity to express their particular views as to what should be included in the data sheets.
		The general -- I would ask you, for the general data sheet, to allow me to present what I've filled out and then go around and ask anybody if they have anything different because it's the more straightforward of the two, and perhaps we can reach a consensus on the first general sheet that way.
		Then, for the supplemental data sheet, we'll ask the FDA to put up the questions, and we will answer -- that is, fill out and come to a consensus on the items on the supplemental data sheet based upon our discussion of the specific questions.
		Dr. Skinner, did you have a comment?
		Oh, I'm sorry, Dr. Besser, I thought that from your lead review that you had had a chance to go around, and I apologize for neglecting you.  Dr. Besser, please accept my apology and offer your comments in the general around the table.
		DR. BESSER:  Your apology is accepted.  It's Dr. Besser.
		Just one comment, I guess, about one of the things that Dr. Skinner had brought up about the fixation of the acetabular shell and the bone/cup interface.  Essentially, as these devices are being designed, you actually do want them to fail either at the liner/cup interface or the femoral head liner interface before they fail at the bone/cup interface.
		So that possibly the published values for the two devices that are currently on the market might be considered more of a target than a goalpost, to steal Dr. Skinner's metaphor.  
		If you make them too hard to either dislocate or disassociate, you're going to have the problems that Dr. Skinner described and risk disassociation at the bone/cup interface instead, which is, as I understand, much harder to correct.
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		I'd like to proceed -- before we go into the second round of the table discussion with the questions, ask once more if we can have an open public session, and ask the persons in the audience if anybody would like to make any comments at this time prior to us going into deliberation on the worksheets.
		Anyone from the audience wish to address the panel at this time?
		Seeing none, let's proceed to the worksheets.  I'm going to go over the general device classification questionnaire first.  I would ask that I be allowed to run through the answers that I've put down on it and then ask if there is consensus on those answers or disagreement.
		The petitioners, the OSMA, the Orthopedic Surgical Manufacturers Association.  The generic type of devices are constrained total hip arthroplasty devices.  
		Question number one, life sustaining or life supporting?  No.
		Number two, is the device for use which is of substantial importance in preventing impairment of human health?  Yes.
		Number three, does the device present a potential unreasonable risk of illness or injury?  No.
		And number four is yes.  Did you answer yes to any of the above three?
		Number seven, is there sufficient information to establish special controls to provide reasonable assurance of safety and effectiveness?  Yes.  If yes, check those controls needed.  
		These would be performance standards, testing guidelines, and in the "other" category, those special controls mentioned by the presenters this morning that can be agreed upon on the supplemental data sheet, and the device labeling controls, again which will be discussed and agreed upon for the supplemental data sheet.
		Number eight, if a regulatory performance standard is needed to provide reasonable assurance of the safety and effectiveness of a Class II or III device, identify the priority for establishing such a standard.  High priority.  No other devices available to deal with this particular patient population.
		Number ten is not applicable.  
		Number 11(a), can there otherwise be reasonable assurance of its safety and effectiveness without restrictions on its sale, distribution or use because of any potentiality for harmful effect or the collateral measures necessary for the device's use?  No.
		11(b), identify the needed restrictions.  In the "other" category, I would suggest that it be limited to use by a surgeon and that the labeling specifications, as will be discussed and agreed upon in a supplemental data sheet, be included.
		I would motion that this be our general device classification sheet and would ask the panel now if anybody has comments relative to it.  If there are some, let me hear now.
		Mr. Dillard.
		MR. DILLARD:  Yes, Jim Dillard.
		I wanted to just make one clarification because it's one that is usually needed for this particular type of exercise.
		Performance standards for number seven, just a point of clarification.  Performance standards here in this context means an FDA developed, mandated through notice and comment rule making performance standard for these particular types of devices.
		Those types of performance standards have generally, in the past, not been very successfully developed.  It's not that we can't do it; it's just that based on notice and comment rule making, they tend to be quite difficult and quite time consuming getting there.
		So in order to check that box, you need to consider that what you're asking FDA to do is develop a mandatory performance standard that will go through that type of process.  Sometimes it's a terminology issue here.
		If you are thinking consensus standards or those standards that have been developed by organizations such as ASTM and ISO, those we would consider to be consensus standards and to be under the "other" category here.
		So, just a point of clarification and one thing that you ought to consider for performance standards.
		DR. YASZEMSKI:  Yaszemski. 
		Mr. Dillard, I was considering consensus standards, and it was my intention that these be voluntary standards by the manufacturer in line with -- manufacturers, that is -- in line with the data that they've presented to us and not anything regulated by the FDA.
		I would change my recommendation, based upon Mr. Dillard's suggestions, and uncheck that box labeled performance standards.
		Other comments from the panel?
		Hearing none, I would suggest that we now go around the table.  
		Dr. Besser.
		DR. BESSER:  Dr. Besser.
		Do we have to answer question nine?
		DR. YASZEMSKI:  Thank you, Dr. Besser.  I neglected question nine and I checked no for question nine.
		Yaszemski.
		MS. SHULMAN:  Actually -- Marjorie Shulman, FDA.
		With taking away the performance standards out of number seven, we don't have to answer eight or nine because --
		DR. YASZEMSKI:  Thank you.
		MS. SHULMAN:  -- questions eight and nine just --
		DR. YASZEMSKI:  Thank you.  So noted.
		Any other comment -- Yaszemski -- any other comments relative to the general questionnaire?
		Hearing none, let us accept this as a draft which we will vote on at a later time in the morning, and let us proceed to the panel questions and the supplemental data sheet.
		Question one.  Question one pertains to block one, generic type of device on the supplemental data sheet.  In addition, question four is going to pertain to block one.  And I would ask then that we perhaps consider question one first and then question four out of order, if we might, and address block one on the supplemental data sheet by our answers to those.
		Question one:  Does the petitioner's proposed classification sufficiently describe constrained hip devices?  If not, what other types of descriptive information should be included in the classification definition?
		And let's go around the table again.
		Dr. Besser, would you mind if we start with you and go around in the clockwise order?
		DR. BESSER:  Dr. Besser.
		I think that the suggested classification, which I don't have in front of me, was -- sufficiently describes the constrained hip devices.  I would like to hear the opinions of some of the orthopedic surgeons as we go around as to whether to include devices that are biologically fixed.
		DR. CHENG:  This is Dr. Cheng.
		I originally thought the answer to this question was yes.  And then, as I thought more, I changed my answer to now because I think there is insufficient evidence to consider use of this in the cemented cup.
		I'm not talking about the femoral stem.  I don't think that's an issue here as much.  And there's -- in answer to Dr. Walker's earlier question, I don't think we can glean much information from that, but it probably doesn't make a lot of difference.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Before we move to Dr. Hannaford, I've asked our FDA colleagues to put up the proposed classification so we can all be looking at it as we're discussing it.
		Dr. Hannaford.
		DR. HANNAFORD:  I don't think this falls within my expertise, so I'll decline to answer.
		DR. YASZEMSKI:  Thank you, Dr. Hannaford.
		Dr. Aboulafia.
		DR. ABOULAFIA:  I do believe that the proposed definition that has been set forth is adequate and don't recommend any changes.
		DR. YASZEMSKI:  Thank you, Dr. Aboulafia.
		Dr. Walker.
		DR. WALKER:  Everything appears adequate to me except for that last sentence, and I still haven't heard any demonstration that with the unusual mechanical loading that this device has characteristics that it has that anyone has shown that biological fixation of the acetabular cup or the femoral component is going to be adequate to prevent failure at that point.
		So, if there's no data, then I don't know -- obviously the biological fixation was not allowed in the original classification and I haven't heard the reason why it should now be allowed in this classification. 
		And I think this is something that has to be done proactively.  We can't just eliminate it without good reason.
		DR. YASZEMSKI:  Thank you, Dr. Walker.
		Dr. Yaszemski here.  For the record, I'm going to read that last statement that Dr. Walker referred to.
		It states, "This generic type of device is intended for use with or without bone cement."
		Thank you.
		Dr. Silkaitis.
		DR. SILKAITIS:  Yes, I guess my comment is in regards to maybe consistency with the previous panel, which I believe maybe the majority of us were here when it was originally reviewed.  I believe that the acetabular components at that time, the indication for the product was use with or without bone cement.
		So if information hasn't changed since then, I would say that the classification as proposed should stand.
		DR. YASZEMSKI:  Thank you, Dr. Silkaitis.
		Mr. Dillard, have you comments?
		MR. DILLARD:  No comments at this time, thank you.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Skinner.
		DR. SKINNER:  Dr. Skinner.
		I think the biological fixation issue is one that's left over from history.  I think that this sentence was put in virtually everything that was put on the market for a number of years, that it's not intended for biological fixation.
		I think that's the reason it was there then and isn't there now.  Maybe Mr. Dillard will correct me on that.  I think the proposed classification is adequate as stated by the manufacturer.
		DR. YASZEMSKI:  Thank you, Dr. Skinner
		Dr. Larntz.
		DR. LARNTZ:  No additional comments.  It looks fine.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.
		Dr. Laurencin.
		DR. LAURENCIN:  Laurencin.
		You know, I think that -- I agree on one hand with Dr. Skinner that that sentence is probably there because it was -- in the time it was being discussed, that sentence was always there in all the different implants.  
		However, for this particular implant, it actually has some special meaning because of the thought that initial fixation is -- can be an issue in terms of dislocation, or actually disarticulation of the implant from the bone.
		So I'm not sure whether it actually should be removed or that we should somehow address it in terms of at least addressing or saying we think it's fine to be in or it should be fine with initial supplementation with screws or some statement to that effect.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		Are there any other comments from the panel on question one?
		MR. DILLARD:  Dr. Yaszemski?  Jim Dillard.
		DR. YASZEMSKI:  Mr. Dillard.
		MR. DILLARD:  If you would like me to address Dr. Skinner's comment or question.  I believe that both Dr. Skinner and Dr. Laurencin are on the right track here.  In terms of this particular statement, when the classification panels met in the last '70s and early '80s, there was no information for biological fixation.
		And I think at the time it was added to pretty much every implant, whether it was on, in the case of a hip, the acetabular side or the femoral side.  It was pretty much in all the comments.
		One of the things this panel has been addressing, I think for all these reclassification petitions, has been the issue about fixation.  And I believe there have been recommendations from this panel on both sides.  Under some circumstances, there's enough information for both biological fixation to be removed, because additional information has come out in the literature in order to support that; and there have also been comments made that there is no information that would support the removal of biological fixation under some of the circumstances.  
		So this particular panel has given us recommendations on both sides and it's been predominantly based on the amount of data that's been available for the particular type of devices.
		There has been also some discussion, I'll just mention, by this panel about whether or not generically the idea of removal of biological fixation could be handled across various joints.  And we have never really had this as an item before this panel.
		In general, it's giving me some good ideas for the future perhaps to do that.  I think in this case I would agree with Dr. Laurencin to say that we would like to have your comments on this particular joint and whether or not -- and this particular implant design, and whether or not that sentence could be removed under this circumstance.
		So with that, unless there's any other questions --
		DR. YASZEMSKI:  Thank you, Mr. Dillard.  Yaszemski.
		I would like to ask FDA to put question four up at this time because the discussion for question one moved over into the topic of question four and I think we've begun that discussion.  And before I make a summary to the FDA of what we feel on this issue, I'd like to ask us to discuss question four.
		Question four is up.  It reads, "The original classification included devices to be fixed with or without bone cement, but excluded devices intended for biological fixation.  What impact does the means of fixation have on constrained designs (for example, cemented, HA-coated, porous-coated or press-fit)?  
		"Has the petitioner provided sufficient information to reclassify devices intended for cemented, uncemented and/or biologic fixation?"
		And I'd like to give everybody on the panel an opportunity to comment again on this because it is posed as a separate question.  And if you feel that your comments to question one have adequately covered your feelings for question four, please so state and we'll move on.
		Dr. Besser.
		DR. BESSER:  Dr. Besser.
		I think that points that have been brought up with reference to question one are -- have already been adequately stated.  I have nothing to add.
		DR. YASZEMSKI:  Thank you.
		Dr. Cheng.
		DR. CHENG:  I do not think the petitioner has provided sufficient information for cemented usage.  For uncemented usage, I think it's insufficient information for a scientist; but reasonably speaking, this is a -- it is a valuable tool sometimes and I think it should be released for use.
		So, from that standpoint, I think it is sufficient.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Dr. Hannaford.
		DR. HANNAFORD:  No additional comments.
		DR. YASZEMSKI:  Thank you, Dr. Hannaford.
		Dr. Aboulafia.
		DR. ABOULAFIA:  I do have one comment.  You know, we're talking about biological fixation and I think in some of our own minds we know that we're talking about primary versus revision when we bring up this issue.
		I think if acetabular component is well fixed, you're revising it, it's biologically fixed, I don't know if any one of us would suggest not to use the constrained liner.  So I think we have to at least make that distinction if indeed you want to make the distinction about biological ingrowth or not.
		DR. YASZEMSKI:  Thank you, Dr. Aboulafia.
		Dr. Walker.
		DR. WALKER:  There are two questions up there.  The first question up there I don't have the expertise to answer.  The impact question, as a scientist, I don't feel I've been provided sufficient information to include biological fixation.
		DR. YASZEMSKI:  Thank you, Dr. Walker.
		Dr. Silkaitis.
		DR. SILKAITIS:  I have not additional comment at this time.
		DR. SILKAITIS:  Thank you, Dr. Silkaitis.
		Mr. Dillard.
		MR. DILLARD:  Nothing further at this time.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Skinner.
		DR. SKINNER:  Well, I think Dr. Laurencin and I are on the same wavelength on this.  I think that in order to put in one of these constrained cups, the interface between the bone and the cup should tolerate tensile stress, should be able to accept tensile stress.
		And some of those up there would, and some of them would not tolerate tensile stress.  And I would want to discourage the surgeon from using one of these constrained cups in a situation that would not tolerate tensile stress because the lever-out mechanism would be likely to pull it out.
		It would be less damaging than pulling out a porous-coated prosthesis, porous-coated acetabulum, but it would still be to the detriment of the patient, I think.
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Dr. Larntz.
		DR. LARNTZ:  I have nothing to add with this particular point, except I do not believe the petitioner has provided sufficient information, period.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.
		Dr. Laurencin.
		DR. LAURENCIN:  Again, I echo what everyone else has said, but the point is that the -- that, number one, I guess we don't have a lot of information in terms of what's in the literature, which I guess was the basis of their petition saying that we do have new information.
		But we don't have a lot of new information about the biological fixation side.  When I looked at the literature, the only paper that actually addressed biological fixation and sort of gave recommendations was Cameron's paper where he actually recommended that for the very initial fixation that all of the ingrowth types of prostheses on the acetabular side should be reinforced with screws, in his experience.
		So my feeling is just with that piece of information I have from the literature is that there should be some reinforcement in the very beginning for initial biological fixation if we're going to be using this implant because we know the stresses are going to be higher than with conventional implants, at least initially.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		Dr. Yaszemski here.
		I'd like at this time, if I can, to provide a summary of the just-completed panel discussion regarding panel questions one and four for the FDA and ask the FDA if we've adequately answered questions one and four.
		It's the view of the panel that the device classification as proposed by the petitioners is such that, from a scientific perspective, not enough information has been presented to cover the indications of cemented use and biologic fixation, especially with respect to the resistance to tensile forces at the cup/bone interface.
		The panel feels that the use of the "not intended for biologic fixation" statement that has previously been included in these types of devices is, in fact, historical, but that scientific information to dispute the non-biologic fixation issue has not been presented.
		On the other hand, if surgeons recognize that these devices should be used infrequently and in relatively limited indications, then these devices constitute a very useful tool in the armamentarium of the surgeon, especially for revision cases, and that perhaps consideration should be given to assuring that there is some resistance to tensile forces thought of and supplied at the time of surgery, in which case these devices will be quite useful to the patients in those limited indications that we've discussed.
		FDA, have we adequately answered your concerns regarding questions one and four?
		MR. DILLARD:  Yes, you have.  I realize that you also have to put some of this down on the supplemental data sheet; but I think from the standpoint of the question, that's adequate.
		Thank you.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Yaszemski here.
		I'd like to ask now that we proceed to question number two.  I'll read question number two.
		"Based on the known clinical information, for which patient populations should constrained hip devices be indicated for use?"
		Dr. Besser, can we start with you again?
		DR. BESSER:  Dr. Besser.
		No comment at this time.
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		Dr. Cheng.
		DR. CHENG:  Well, I would say patients who have recurrent dislocations which are not solvable or amenable to any other solution.  I mean, this is only to be used when the surgeon's back is up against the wall, and we discussed one of those positions or the components.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Dr. Hannaford.
		DR. HANNAFORD:  No comments.
		DR. YASZEMSKI:  Thank you.
		Dr. Aboulafia.
		DR. ABOULAFIA:  I think it's a reasonable question.  In some ways, it becomes a difficult one to put down on paper to answer, only because Dr. Skinner suggests that something like "other methods likely to fail" certainly works better than some of the other ones that have been proposed.
		But I think ultimately it becomes a clinical decision.  You might opt for a 45-minute operation and someone who does have alternative methods of solving the problem, but those alternative methods may be a four hour operation that you didn't think is in the patient's overall best interest.
		You know, in the example of the dislocated constrained prosthesis that Dr. Skinner showed, he showed an acetabular component that looked clearly malpositioned.  But I think there are cases when you're looking for a quicker solution; specifically those very low-demand, high-risk, high operative morbidity patients.
		So I think when you actually try and put it down into words, you have to leave some leeway for clinical decision making.
		DR. YASZEMSKI:  Thank you, Dr. Aboulafia.
		Before we ask Dr. Walker for his input, let me read for the record the proposed indications for uses.  This is Dr. Yaszemski.
		"Patients at high risk of hip dislocation due to a history of prior dislocation, bone loss, soft tissue laxity, neuromuscular disease or intraoperative instability."
		Dr. Walker.
		DR. WALKER:  No further comments.
		DR. YASZEMSKI:  Thank you.
		Dr. Silkaitis.
		DR. SILKAITIS:  Yes, the proposed indications is very similar or almost identical to the one that was reviewed a year ago, so that's acceptable.
		DR. YASZEMSKI:  Thank you, Dr. Silkaitis.
		Mr. Dillard.
		MR. DILLARD:  No comments.
		DR. YASZEMSKI:  Thank you.
		Dr. Skinner.
		DR. SKINNER:  Harry Skinner.
		I only wanted to add, as I said in my previous presentation, and in whom other -- 
		MR. DEMIAN:  Excuse me.  We're having trouble hearing Dr. Skinner over here.  Like perhaps his microphone's not working?
		DR. ABOULAFIA:  Aboulafia.  
		Other methods likely to fail.
		DR. SKINNER:  Yes, that's it.  
		Thank you.
		(Laughter.)
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Dr. Larntz.
		DR. LARNTZ:  The only comment is that there's not much clinical information to change anything and it's clear that this device has to be used sparingly given the lack of clinical information.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.
		Dr. Laurencin.
		DR. LAURENCIN:  I have to agree with Dr. Aboulafia.  I wouldn't want to tie the orthopedic surgeon to having to be second-guessed at the end as to whether there was another method that might have been possible to work, but may have taken, say, ten hours to perform in a patient that may have a revision after some metastatic tumor where the original -- a lot of bone loss where the original operation was done and the patient -- they want to get the patient off the table for the patient's health.
		So I think that some midway has to come in where -- a midpoint has to be achieved where perhaps all other options have been considered and perhaps some language where all options have been considered and this option has been deemed best for the particular patient.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		Dr. Yaszemski here.
		I'd like to -- Mr. Dillard, I'm sorry.
		MR. DILLARD:  Jim Dillard.
		I thought I might just give you a process option in this case that has been used in the past.  If, contained within the indications for use, you want to give a little bit more flexibility, one of the ways that we have handled that is to either include warnings, other contraindications if so noted, or precautions that might be additionally helpful to a surgeon that can also be used as special controls in order to get a point across if you believe that it would be appropriate for all of the products of the category type.
		So I thought I'd just lay out as a potential option that you might want to consider.
		Thank you.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Yaszemski here.  
		I'd like to summarize now the panel's discussion regarding question two.  We feel that the proposed indications as listed are appropriate, with the additions suggested by Drs. Aboulafia and Skinner, of a transmission to the surgeon that all other non-constrained options should have been considered or are likely to fail.
		With this, to FDA, have we adequately answered question two for you?
		MR. DILLARD:  Yes, thank you.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Let us proceed now to question three.  Question three relates to item number five on the supplemental data sheet.  
		Question number three reads, "Risks to health have been identified by the petitioner, previous panel and the medical device reports.  Have all the risks to health for a constrained hip prosthesis been identified?  If not, what additional risks should be described?"
		Dr. Besser.
		DR. BESSER:  No comment at this time.
		DR. YASZEMSKI:  Thank you.
		Dr. Cheng.
		DR. CHENG:  I think they've been identified.
		DR. YASZEMSKI:  Thank you.
		Dr. Hannaford.
		DR. HANNAFORD:  No comment.
		DR. YASZEMSKI:  Thank you.
		Dr. Aboulafia.
		DR. ABOULAFIA:  Agree with Cheng.
		DR. YASZEMSKI:  Thank you.
		Dr. Walker.
		DR. WALKER:  It's a very comprehensive list.  Nothing needs to be added.
		DR. YASZEMSKI:  Thank you.
		Dr. Silkaitis.
		DR. SILKAITIS:  No additional comment.
		DR. YASZEMSKI:  Thank you.
		Mr. Dillard.
		MR. DILLARD:  No additional comments.
		DR. YASZEMSKI:  Thank you.
		Dr. Skinner.
		DR. SKINNER:  No additional comments.
		DR. YASZEMSKI:  Thank you.
		Dr. Larntz.
		DR. LARNTZ:  No comment.
		DR. YASZEMSKI:  Thank you.
		Dr. Laurencin.
		DR. LAURENCIN:  Agree with the above.
		DR. YASZEMSKI:  Thank you.  
		Dr. Yaszemski here.
		I'll summarize then that the panel is in agreement that all the risks to health have been adequately identified and no additional risks or cautions need to be added.
		To FDA, Mr. Dillard, have we adequately answered question four for you?
		MR. DILLARD:  Yes, you have.  Thank you.
		DR. YASZEMSKI:  Thank you.
		We've now discussed all the questions that the FDA has posed to us, and we now come to the task of proposing answers -- that is, filling in of the general worksheet and the supplemental worksheet.  And I would ask at this time whether anyone on the panel has a specific wording that they feel would be appropriate for the supplemental data sheet.
		If not, I'll try to summarize it from the discussions that we've had, and then we'll ask if there might be a motion to accept the worksheets as filled out.  
		With respect to item one on the worksheet, the generic type of device, we feel that, in addition to the description proposed by the petitioners, it would be perhaps useful to include a notice to the surgeon, a warning to the surgeon that there is not scientific evidence that supports effectiveness in cemented or biologic fixation instances and that perhaps some consideration should be given to providing immediate resistance to tensile forces at the time of insertion.
		Do I have comments from the panel regarding that wording and whether folks feel it's appropriate or inappropriate?
		Dr. Cheng.
		DR. CHENG:  Well, I might disagree with the wording that the FDA's acting on something without any scientific evidence.
		DR. YASZEMSKI:  May I ask for a proposal as to how it would be most effectively reworded?
		DR. CHENG:  I might indicate that there is limited experience in the use of these devices, and therefore they be considered for usage only when all other options have been exhausted.  
		And then, I might add, unless clinical indications indicate otherwise, or suggest otherwise -- such as the cases that Dr. Aboulafia mentioned and so forth.
		DR. YASZEMSKI:  So other comments on that?
		Would it be appropriate, Dr. Cheng, to add some statement about consideration for immediate resistance to tensile forces as Dr. Skinner has mentioned?
		DR. CHENG:  Yes, I would agree to that.  But I might also bring up -- Dr. Skinner and Dr. Laurencin, are you referring to putting screws into cups when the cup is originally put in as a revision case, or what about an existing, well fixed, ingrowth cup and you go to change the liner; do you think you need to add screws or not add them?
		I don't know the answer to that question.
		DR. YASZEMSKI:  Dr. Skinner.
		DR. SKINNER:  Well, again, I was trying to be vague on the type of fixation that we're talking about.  Anybody who has removed one of the relatively smooth coated, hydroxyapatite-coated cups knows that at revision it's a wonderful revision to do because the cup just pops out.  It's very nice from that viewpoint.  
		In those cases, there's no resistance to tension.  In that situation, if the cup doesn't come out or is going to be difficult to come out for some other reason, then perhaps an additional screw or two might be necessary to resist the tension.
		If it comes out easily, then it would be easy to put in another cup.  Some of the cups -- for instance, the beaded cups, Howmedica, the fiber metal from Zimmer, etc. -- those, if bone grows into them, will provide a tensile load, a tensile resistance that, after their ingrowth, there won't be any problem with having additional screws.
		That will take care of it in that situation with or without screws.  Those are hard to get out.
		DR. LAURENCIN:  I'm mainly referring to initial fixation for --
		DR. YASZEMSKI:  Excuse me, Dr. Laurencin speaking.
		DR. LAURENCIN:  Dr. Laurencin, I'm sorry.
		I'm referring really to initial fixation of the cup than a new cup being placed in the location.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		Other comments on the generic type of device?  Let's move to number four, indications, on the supplemental data sheet.  The proposal we have is for that recommended by the petitioner with the addition of "all other options to constrained hips have been considered and deemed not appropriate."
		Comments on this, which will be number four?
		Dr. Skinner.
		DR. SKINNER:  Just one.  I would change hips to acetabular components.
		DR. YASZEMSKI:  Thank you.  So noted.
		Other comments on number four?
		DR. ABOULAFIA:  Aboulafia.
		You said all other options have been considered and deemed inappropriate.  Do we need the "and deemed inappropriate?"
		DR. YASZEMSKI:  It would be fine with me to just say considered. 
		Yaszemski.
		Thank you.  We'll move to number five, identification of any risk factors presented by device.  I propose that we fill this in as proposed by the petitioner.
		Comments on this?  
		With that, I believe we've given consideration and have come to consensus on what needs to be included in the general classification sheet and the supplemental data sheet.  Might I ask if there's a motion for reclassification at this time from the floor?
		Dr. Skinner.
		DR. SKINNER:  Dr. Yaszemski, I would like to move that we recommend reclassification to Class II for these type of devices.
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Is there a second to this motion?
		DR. LAURENCIN:  Laurencin to second.
		MR. DILLARD:  Excuse me.
		DR. YASZEMSKI:  Mr. Dillard.
		MR. DILLARD:  Yes, Dr. Yaszemski, a point of process.  If you wouldn't mind, I think it would be helpful to actually go through the entire supplemental data sheet and have some verbiage for the record that we can make sure that we fill in there before moving towards a vote.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		MR. DILLARD:  Thank you.
		DR. YASZEMSKI:  Mr. Melkerson, could I ask you to put those up that we discussed, and I'll suggest that I'll read them as they're put up and be certain that everybody agrees with them.
		For the supplemental data sheet under number one, we're going to say constrained total hip arthroplasty devices.  The additions that we discussed for number one we will put down in number nine, identification of any needed restrictions on the use of the device.
		And we'll include -- Mr. Melkerson, may I ask you to put the proposed classification one up in which we put the verbiage on the bottom so that I can read that as the number nine?
		We are going to add on one of the proposed restrictions that the surgeon should consider providing immediate resistance to tensile forces in the case of initially fixed biologic acetabular cup -- biologic fixation acetabular cups.
		Thank you, Mr. Melkerson.  May I ask you to put up the indications as put forth by the petitioner?
		This will be number four on the supplemental data sheet.  It will be as proposed by the petitioner with the addition that "all other options to constrained cups have been considered."
		And we'll change hips to cups there, Mr. Melkerson, please.  And under number five, we will fill the supplemental data sheet in with "risks as proposed by the petitioner."
		Mr. Dillard, does that adequately answer your verbiage question?
		MR. DILLARD:  Yes.  And if you wouldn't mind continuing through the sheet -- Mark, if you would put up supplemental data sheet.
		DR. YASZEMSKI:  Thank you.  Thank you, Mr. Dillard.
		MR. DILLARD:  Could I make one other point of order, Mr. Chairman?
		DR. YASZEMSKI:  Yes.
		MR. DILLARD:  Just to remind all the panel members to make sure that they are filling out their own sheet because we will collect it at the end.  
		Thank you.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Mr. Melkerson, we'll go through specific hazards to health or as proposed.  Number six, recommended advisory panel classification, the motion and the second are for classification into Class II and with high priority.
		Number seven, I would propose we say "as presented by the petitioner."  
		And for number eight, "as presented by the petitioner."
		In number nine, Mr. Melkerson, I would ask you to add the statement that we put on the petitioner's classification proposal with respect to providing resistance against tensile forces.
		As Mr. Melkerson is filling this out, I would ask, Dr. Skinner, is this an adequate representation of your motion?
		DR. SKINNER:  Yes.
		DR. YASZEMSKI:  And Dr. Laurencin, the motion that you seconded?
		DR. LAURENCIN:  Yes.
		DR. YASZEMSKI:  Thank you.
		It has been moved and seconded that the constrained total hip arthroplasty devices be classified into Class II as outlined on the supplemental data sheet that we have in front of us.  I'm going to go around the table now and ask --
		MS. SHULMAN:  Excuse me.
		DR. YASZEMSKI:  Yes.
		MS. SHULMAN:  Marjorie Shulman, FDA.
		DR. YASZEMSKI:  Ms. Shulman.
		MS. SHULMAN:  There's a back part to the supplemental data sheet.
		DR. YASZEMSKI:  Thank you, Ms. Shulman.
		MS. SHULMAN:  It's quick.
		DR. YASZEMSKI:  Number ten, the device is not Class I, so it's not applicable.
		MS. SHULMAN:  Also, as a matter of housekeeping, these forms have to be updated, but you can vote for a Class II to be exempt from 510(k).
		DR. YASZEMSKI:  Thank you.
		Number 11, existing standards applicable to the device, device of assemblies, or device materials.  I would ask FDA if there's any particular information that we need to put in this?
		MR. DILLARD:  Jim Dillard, FDA.
		You may reference the petition here, as you've done in others.  And then, if there are any others you would like to reference that are not included in that list, we could note them here.
		DR. YASZEMSKI:  Yaszemski.
		I would suggest that we state for number 11 "as per the petitioner's proposal."
		Dr. Besser.
		DR. BESSER:  Dr. Besser.
		I'd like to also include reference for the preclinical testing to include stem types of the type that Dr. Skinner was speaking of where significantly lower pull-out -- or higher pull-out stresses would be expected because of the stem design.
		DR. YASZEMSKI:  Dr. Yaszemski here.
		Mr. Dillard, the petitioners mentioned testing that has been done.  Would it be such that we could include Dr. Besser's recommendation in the testing information as put forth by the petitioner without making it a specific requirement for approval?
		MR. DILLARD:  Jim Dillard.
		Yes, you may do that.  It also helps just with the discussion for FDA to note it, and it may be one of those things that we would look for in review of a 510(k).  So it does not need to be specifically mentioned as a standard here, but to note for us to take a look at that when we're looking at differential stem designs.
		DR. YASZEMSKI:  I would propose then -- and after making this, I'll come back to Dr. Besser to ask if it meets his approval -- that we would not make it a requirement for the approval, but would recommend to FDA, based upon the discussions we've heard from Dr. Besser, Dr. Skinner, and Drs. Brown and Brooker in the presentation regarding lever-out, pull-out, range of motion and arc of motion, that we would make a recommendation to the FDA that they should include these things and recommend to the manufacturers that they come to agreement on appropriate values for these and give consideration to them as a group, but not to make them an absolute requirement for reclassification into Class II.
		Dr. Besser, would that satisfy you?
		DR. BESSER:  Dr. Besser.
		Yes, that would be satisfactory.
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		And Mr. Dillard will -- we will say "as proposed by petitioner without including testing to address stem types.  And we'll make that a recommendation to you, but not part of the criteria for reclassification.
		Dr. Larntz.
		DR. LARNTZ:  Kinley Larntz.
		Just to make sure we're perfectly clear, the two approved devices have apparently very different behavior when they're actually used out in the world, at least according to our limited data.  And we want to make sure -- I just want to make sure everyone's aware that it looks like the two approved devices are quite diverse with respect to, for instance, dislocation rate, which is the primary item that we had information on.
		I don't know that there's anything that needs to be done about that, but I want the panel to be very aware of that and go from there.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.  So noted.
		Dr. Yaszemski here.
		I'll restate that we have a motion to reclassify constrained total hip arthroplasty devices -- and a second to that motion -- into Class II with the particular specifications as outlined on the supplemental data sheet. 
		And I would like to go around the room now and ask each panel member to provide their vote, yes or no, for this motion as it appears on the supplemental data sheet, and to offer a reason for their vote.
		Dr. Besser, we've been starting with you.  May I ask you again to begin?
		MR. DILLARD:  Dr. Yaszemski?
		DR. YASZEMSKI:  Mr. Dillard.
		MR. DILLARD:  Thank you.  I hate to intervene.
		DR. YASZEMSKI:  Please do.
		MR. DILLARD:  A point of clarification in process again.  Just to make sure that everybody understands, we have to vote on both of the sheets, both the supplemental data sheet and the original sheet.
		It might be helpful to take them in the opposite order in which you want to take them.
		DR. YASZEMSKI:  Dr. Yaszemski or Mr. Dillard, should we vote twice then?  Vote twice?
		MR. DILLARD:  Yes, please.
		DR. YASZEMSKI:  Would it be appropriate from the FDA's perspective to begin the vote with the supplemental data sheet vote?  Or should we vote with the general first?
		MR. DILLARD:  Jim Dillard.  I believe it would be better to start with the general device classification questionnaire and then go to the supplemental data sheet.
		DR. YASZEMSKI:  Mr. Melkerson, could I ask you to put the general device questionnaire up so that we can look at it.
		All again, we will call for a vote.
		We will go around the room twice,  Yaszemski here, and vote first on the general device classification sheet as it appears in front of us.
		I will start with Dr. Besser again.
		DR. BESSER:  I vote for the motion based on the discussion of the past two hours.
		DR. YASZEMSKI:  Dr. Cheng?
		DR. CHENG:  I vote for approval.
		DR. YASZEMSKI:  Dr. Hannaford?
		DR. HANNAFORD:  I am going to abstain entirely due to my own lack of expertise on this topic.
		DR. YASZEMSKI:  Dr. Aboulafia?
		DR. ABOULAFIA:  I will vote for approval.
		DR. YASZEMSKI:  Dr. Walker?
		DR. WALKER:  I vote for approval.
		DR. YASZEMSKI:  Thank you.  Dr. Skinner?
		DR. SKINNER:  I vote for approval.
		DR. YASZEMSKI:  Dr. Larntz	?
		DR. LARNTZ:    I vote for approval.
		DR. YASZEMSKI:  Dr. Laurencin?
		DR. LAURENCIN:  I vote for approval.
		DR. YASZEMSKI:  The vote is seven yes and one abstention.  The motion passes for the general data sheet.
		We will now move to the supplemental data sheet.  I would ask Mr. Melkerson to put it up again.
		This vote will be for the supplemental data sheet, as filled out.
		Dr. Besser?
		DR. BESSER:  Dr. Besser.  Yes.
		DR. YASZEMSKI:  Dr. Cheng.
		DR. CHENG:  I vote to approve.  No further comments.
		DR. YASZEMSKI:  Dr. Hannaford?
		DR. HANNAFORD:  Abstain, as before.
		DR. YASZEMSKI:  Dr. Aboulafia?
		DR. ABOULAFIA:  Approve, as is.
		DR. YASZEMSKI:  Dr. Walker?
		DR. WALKER:  Yes.
		DR. YASZEMSKI:  Dr. Skinner?
		DR. SKINNER:  I vote yes.
		DR. YASZEMSKI:  Dr. Larntz?
		DR. LARNTZ:    Yes.
		DR. YASZEMSKI:  Dr. Laurencin?
		DR. LAURENCIN:  I vote for approval.
		DR. YASZEMSKI:  The vote is seven yes, one abstention, and the motion for the supplemental data sheet passes.
		FDA.  The recommendation of the panel is that the general data sheet and supplemental data sheet, as presented to you for reclassification into Class II of Constrained Total Hip Arthroplasty Devices has passed, and we recommend to you that they be classified as Class II devices.
		Mr. Dillard?
		MR. DILLARD:  Yes, Dr. Chairman.  Could I ask for one more, very quickly, could you just go around and ask people to state for the record what their reasons were for an approvability vote or abstention, in terms of the two sheets?
		Thank you.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Besser?
		DR. BESSER:  Dr. Besser.  As previously stated in the discussion of the last two and a half hours, I think this is an important product to be made available to orthopedic surgeons in those limited situations where they are going to need it.
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		Dr. Cheng?
		DR. CHENG:  I think I would tell the FDA that I think it is a useful device in limited situations to be considered when other means for dealing with recurring dislocation have either been exhausted or are not indicated, due to the patient's clinical condition.
		My only concern about the reclassification of this product is that surgeons will have a very low threshold to suddenly reaching for the shelf and using this, perhaps when it is not in the patient's best interest, but because the surgeon feels that it is the easiest way out of a very difficult situation, and then it might be used inappropriately.
		So, I would want to try to prevent that.  I think that is the general feeling that I heard this morning, in putting in some of these safeguards.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Dr. Hannaford, might I ask for your reasons for the two abstention votes?
		DR. HANNAFORD:  I will briefly elaborate.
		The abstention is not meant to reflect on either the devices in question or the process that is going on here.
		It is just the fact that I don't feel my own knowledge on this topic is sufficient to give a quality vote in either direction.
		DR. ABOULAFIA:  I would summarize the discussion really, as --
		DR. YASZEMSKI:  Excuse me, this is Dr. Aboulafia.
		DR. ABOULAFIA:  I would summarize, and to paraphrase someone else's words, it is a simple solution to a difficult problem.
		And to summarize Dr. Cheng's remarks, you don't get something for nothing.  And I think we achieved the goals that we intended to set out.
		DR. YASZEMSKI:  Thank you, Dr. Aboulafia.
		Dr. Walker?
		DR. WALKER:  I think Dr. Besser and Dr. Cheng have both given exactly the same reasons that I would give, and that is why I voted yes.
		DR. YASZEMSKI:  Thank you, Dr. Walker.
		Dr. Skinner?
		DR. SKINNER:  I agree with what has been said.  I think that the discussion and the data provided by the petitioner were adequate to verify the validity of the conclusion we have come to.
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Dr. Larntz?
		DR. LARNTZ:    I actually think that the information provided by the petitioner was inadequate, but I certainly appreciate the expertise of the panel members who provided adequate information for reclassification.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.
		Dr. Laurencin?
		DR. LAURENCIN:  I have nothing more to add.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		One housekeeping item before we adjourn and that is that we finished a bit ahead of schedule, but we are going to need to stick to the afternoon schedule and start at 1:30 p.m.
		However, I would ask everybody to please be back in plenty of time so that we can actually start at 1:30, rather than just start assembling at 1:30.
		With that, we will conclude the morning session and adjourn.
		(Whereupon, the morning session adjourned at 11:41 a.m.)




	A-F-T-E-R-N-O-O-N  S-E-S-S-I-O-N
	(1:31 p.m.)
		DR. YASZEMSKI:  Well, now, I think we are ready to begin.
		May I have your attention please?  We are going to begin the afternoon session.  May I ask everybody to take their seat and we are going to get started at this time?
		We will now proceed with the open public hearing session of this meeting.
		I would like to ask at this time that all persons addressing the panel come forward and speak clearly into the microphone, as the transcriptionist is dependent on this means of providing an accurate record of this meeting.
		We are requesting that all persons making statements during the open public hearing of the meeting, disclose whether they have financial interests in any medical device company.
		Before making your presentation to the panel, in addition to stating your name and affiliation, please state the nature of your financial interest, if any.
		We have one group wishing to address the panel, and at this time I would like to invite Christina Gabriel, president and CEO of CASurgica, Inc. to provide her comments.
		DR. GABRIEL:  Good afternoon.  My name is Christina Gabriel, and I am the new, as of a month ago, president and CEO of CASurgica, which is a very small company in Pittsburgh.
		The company was founded by an orthopedic surgeon and a civil engineer.  Anthony DiGiola is an orthopedic surgeon and the engineer is Branislav Branco Jaramaz.
		They founded the company in 1997 to follow on from over six years of research that has been done collaboratively between Carnegie Mellon University's Robotics Institute and UPMC Shadyside Hospital in Pittsburgh.
		I should state, therefore, that I do have a financial interest in this company.  The company doesn't yet have a product, but we intend to have a product sometime, and it will probably be in this field.
		So, this meeting today is timely and of great interest to us.
		Thank you very much for the opportunity to make a brief statement about issues that we believe that the regulatory process should consider in evaluating devices and technologies for computer-assisted surgery.
		Our perspective on these issues has been developed during more than six years, as I said, of research at the Carnegie Mellon University Robotics Institute and University of Pittsburgh Medical Center, Shadyside Hospital in Pittsburgh, Pennsylvania.
		At these research centers, orthopedic surgeons collaborate closely with researchers in computer science and engineering to develop advanced surgical tools and technologies that will hopefully improve patient's outcomes.
		As part of this collaboration, Shadyside Hospital maintains what they call the Total Joint Registry which includes a general clinical and radiographic data base for patients undergoing total hip or total knee replacement surgery and knee arthroscopies to facilitate the evaluation of joint reconstructive procedures.
		Patients are evaluated pre-operatively as well as post-operatively at three months, six months and annually thereafter.
		As part of the research program, an image-guided surgical planning and navigation system for total hip replacement surgery is undergoing a clinical trial at the hospital with about 100 total hip replacement procedures having been performed to date, using the computer-assisted system.
		Our statement really is as follows; we really have pretty much one point to make.
		The surgical goal, as all the surgeons in the room know, is to enable the patient to recover as fully as possible as quickly as possible with as few complications as possible.
		Current surgical practice is a loosely connected and sometimes uncoupled sequence of events.  Diagnosis and planning, surgical execution of the plan and monitoring of the patient's recovery over time.
		All of us would no doubt agree that the reason we are developing these new technologies is so that the surgeon will be able to accomplish the surgical task and achieve the surgical goal more successfully than is possible using current, unassisted surgical practice.
		Therefore, we believe that one of the key factors that the regulatory process should take into account in evaluating any system designed to assist surgical interventions is the level of control maintained by the surgeon.
		There is a broad spectrum of available and proposed technology, from passive systems to semi-active systems to fully active or robotic systems.
		We would define passive systems as those that provide the surgeon with additional information prior to and during a procedure, but do not perform an action.
		Active systems are capable of performing individual tasks or entire procedures autonomously.
		In between these two extremes, semi-active systems are ones in which the surgical actions are constrained by a robotic system but the surgeon remains in control.
		Our research program has emphasized collaboration between computer scientists and engineers who understand what the technology can do well, and orthopedic surgeons who understand what trained and experienced humans do well.
		We think a good design for these systems is one in which the machine's capability is coupled with human judgement and skill in order to perform a task better than either could do alone.
		The systems should be designed from the surgeon's point of view so that it is easy to use as a part of the normal flow of the surgical procedure.
		Passive computer-assisted surgery provides the surgeon with richer information to draw upon during pre-operative planning and the procedure itself, but leaves all decision-making and control to the surgeon.
		Therefore, we believe that the safety considerations for patients, when such passive systems are used will be significantly different from those associated with active systems that replace any of the surgeon's traditional or typical actions, at any point.
		That is really all we wanted to say today.
		Thank you.
		DR. YASZEMSKI:  Thank you, very much.
		Do we have anyone else who would like to address the panel, at this time, from the public?
		Seeing no hands, we will now proceed to the open public hearing session regarding the development of computer-controlled surgical systems, designed for use in orthopedic procedures.
		First the FDA will present their chosen points and questions.  This will be followed by the lead panel reviewers and then we will have a general discussion.
		I would like to begin by asking Mr. Neil Ogden, branch chief of the general surgical branch, to provide the FDA presentation and questions.
		Mr. Ogden?
		MR. OGDEN:  Thank you, Mr. Chairman.
		(PAUSE)
		Thank you.
		I am Neil Ogden and I am branch chief for the general surgical devices branch, here at the FDA.
		I have a little cartoon here which actually was talked about by Dr. Gabriel very well; thank you for that.
		The little cartoon there on your right is where we would like to see patients; healthy, fit, physically active.  As surgical procedures have been developing and surgical tools, as we will see here today, they have been getting increasingly more complicated.
		Often times now there is pre-op scanning, imaging, and that information is then used via computer systems and software to be mapped on to patient's anatomies.
		That information is then taken into the surgery and hopefully it will facilitate rapid healing and recovery.
		I am first going to talk a little bit about the history of these devices in the agency, then discuss technology a little bit, then some of our concerns and then we will go over the questions we have provided to the panel.
		Pre-70's and in the 1970's typically surgical devices consisted of manual clamping systems, sometimes it was basically a ring attached to a surgical table and clamps and manipulators were then screwed down onto that to hold them in place.
		These could hold various clamps, scopes, retractors.  An example would be the Iron Intern or a Brookwalter clamp.  Pretty simple technologies, easy for an engineer like me to understand, forces loading.
		Through the 1980's the companies started to develop gas-powered arms where various gases are used to control locking joint mechanisms.  Sometimes these were fairly complex, sometimes simple.
		Then in the 1990's, companies started to integrate computer systems and sophisticated software.  They started using sophisticated motorized systems with feed-back loops, incorporating memory for surgical tools and arms, putting on actuators.
		They were also doing a lot of pre-op planning, and using software to map the pre-op anatomy of the patient onto a real-time anatomy of the patient, using that to facilitate a surgical procedure.
		So, going from the simple mechanical systems, now we have the technology today that consists of computer-assisted, which involves the software, the computer hardware, monitors and control interfaces which could be touch-screens or could be voice-activation, could be hand-held pendants.  We have seen a lot of different scenarios.
		Often times it is using pre-operative planning, various imaging modalities, MRIs, CTs, ultrasound is now starting to be used and that is being digitized.
		It is taken into the OR and used to overlay on the patient during real-time procedures.
		Oftentime this also incorporates databases of implant specifications.
		And last but not least, robotics.  Companies are developing systems now that actually assist in performing the procedures as well.
		Here is another kind of categorization of what these technologies are.  This is sort of hierarchical because the systems on the bottom also typically incorporate all the ones that came before it.
		Dr. Gabriel talked about her system of describing these, and that is a good one as well.
		Typically, the first category, computer-assisted retractor holder, an example was given in your panel packs of the Robotrac system.  It is a fairly simple retracting device that surgeons have used in orthopedic procedures.
		Then computer-assisted operative planning machines.  Actually, Dr. Gabriel's group has been developing one of these.  I believe it is called the Hip-Nav System with Dr. DiGiola.
		This machine is used with the pre-operative imaging and planning, and sort  of maps out the best positioning of the acetabular cup.
		Then progressing further to more complicated systems.  You have systems that then provide the pre-operative planning, the computer analysis.
		Then when you take them into the OR, like the Hip-Nav as well, they actually provide some kind of physical guidance to the surgeon.  Either an alignment mechanism or something of that nature.
		The most sophisticated systems from our point of view are the computer-assisted operative planning systems that also include surgery performance, where there is a robotic or some type of motorized mechanism tied into the computer software that actually performs part of the surgical procedure.
		FDA's concerns about these technologies' risks, could be a technical failure of some kind, and how does this transfer into a risk to the patient.
		There could be additional safety issues regarding the use of this technology.
		For instance, does it take longer to do it?
		Does incorporating this kind of technology into the procedure, does it add a lot of additional steps and increase the difficulty of the procedure?
		Also, clinical outcome is very important.  Does the use of this technology improve the clinical outcome, does it make it about the same but add increased risks during the procedure or is the clinical outcome a little worse than traditional methods?
		There may be other concerns as well, as far as risks.
		Benefits.  Well, there could be improved clinical results.
		There could be improved safety profiles.  Using a computer-assisted system may provide the surgeon with enough information to allow them to perform the surgery in a more safe way, perhaps quick, perhaps better aligned.
		Also, surgeon preferences, having a computer-assisted system may make the procedure much simpler for the surgeon because they may not have to spend as much time during the procedure doing their own alignments, assessing what size prosthetic they need to implant.
		There may be other concerns as far as benefits, as well.
		I apologize for the slide not fitting all the way on the screen.  There are numbers, if you follow along.
		So, our concern for the panel is to try to help the FDA understand what types of information we really need to adequately assess these types of technologies.
		Our first question is, please discuss the types of issues and engineering concerns that would be important to evaluate these technologies.
		Mr. Chairman, I am not sure if you want me to read all the questions or do them one at a time and have you respond?
		Read them all, then respond later?  Okay.
		Number two is, please discuss important clinical study endpoints to consider for evaluation of these types of devices.
		Please discuss any longer term safety concerns that need to be addressed in the study of these devices.
		Number three, regarding surrogate endpoints for a computer-assisted surgical technology, are there quantitative and/or qualitative short-term endpoints that could best capture an improvement in the procedure?
		Number four, what longer-term effectiveness endpoints, including clinical endpoints, would be important to consider in looking at risk-benefit for these products?
		Thank you.
		DR. YASZEMSKI:  Thank you, Mr. Ogden.
		We are going to have a general panel discussion aimed at providing FDA with our recommendations regarding these four questions.
		I would like to begin this discussion by having our two lead reviewers present their positions.
		First, I would ask Dr. Walker to lead off the panel's discussion with his pre-clinical review.
		Dr. Walker?
		DR. WALKER:  Mr. Chairman, it is my understanding that there is one review in the open session and one review in the closed session, am I right?
		DR. YASZEMSKI:  Correct.
		DR. WALKER:  So, I think the second review will be delayed.
		MR. DILLARD:  Mr. Chairman?
		DR. YASZEMSKI:  Mr. Dillard.
		MR. DILLARD:  Yes, in terms of the open session, I might say something just for the ground rules here is that the open session is for the public, and what is said here is for the general public.
		The closed session for this afternoon which will be immediately after the open session, will not be for the general public.  At this point, no discussion should ensue about what will be discussed in the closed session.
		This is strictly an open session with general discussion about these topic areas.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Walker, does that answer your question?
		DR. WALKER:  Yes.
		DR. YASZEMSKI:  Please proceed.
		DR. WALKER:  Neil, in his presentation immediately before this one from FDA, in the hierarchy of different levels of computer-assisted surgical devices, had as the highest and most complex device one that was a computer-assisted planning and robotic performance of the surgery.
		And that is the issue from an engineer's point of view that I would like to address.  What I would like to do is lead off for the first part of this into the question of what are the issues and engineering concerns surrounding any sort of a robotic surgical device, and since I am an engineer, this happens to be an engineer's view.
		It happens that I live in New Orleans, and blizzard season hit last week when it got down to fifty degrees.  So, my wife asked me to make a peg board that could go underneath the stairs there the kids could hang up their jackets because they really needed those jackets.  It didn't even get above sixty one day.
		So, my son and I went down to the workshop and grabbed an old mop handle and a piece of 1 x 4 wood, and decided to make some pegs and put them in the wood.
		There is a reason that I am telling you this story that will come out in a minute.
		My younger son is ten, and he decided to do this with a hammer and a screw driver.  His approach to putting a diagonal hole into the board he was going to put the peg into was to get the hammer, tap on the screw driver and make a hole that way, in this 1 x 4 piece of wood.
		My approach, and he was scared to use the electric drill.  My approach was just to use an electric drill.  But as any of you who have ever done that knows, it is awful hard to drill a hole on a diagonal with an electric drill.
		What I really should have done if I had wanted to do this with some precision would have been to go over the lab at school where there is a drill press, and clamp the board to the drill press, set up the correct angle, and have a mechanical device track the bit.
		So, there were three different approaches to putting a hole in a piece of wood and putting a peg in that piece of wood.
		The three approaches were banging on it, drilling it by hand or using some sort of a controlled device.  The controlled device would of course given me much greater three-dimensional accuracy because I would have been drilling on an angle properly, and I could have, in fact, drilled multiple holes at exactly the same angle, and I wouldn't have gotten as tired as I was fighting the electric drill, my precision would have been higher.
		Fortunately, for this particular application, I didn't need to do anything inside the hole after I drilled it, except put glue in it.
		But had I needed to, the robotic approach would have given me that ability.  And I certainly didn't need to see inside the hole, although there are many applications, you can imagine, where that direct visualization of the inside of the hole would have had some great benefits for me.
		So, the analogy between drilling a hole in a piece of wood and doing a robotic surgery is not that farfetched.
		In order to drill that hole repeatedly, I need to introduce the concept that is familiar to the engineers here, and I beg your indulgence, and not so familiar to the surgeons, of closed-loop feedback and measurement.
		Closed-loop feedback and measurement is predicated on the assumption that you want to know where you are, where you are going, and that an automated device figures out how to get from where you are to where you want to be.
		Drilling a hole, I know that I am at the top of the piece of wood and that I want to go through to the bottom of the piece of wood.
		Only now we will add into that a measurement of where I am, some sort of a automatic device that says where I am, figures out how fast to advance the drill press, and apply a correction so that the drill goes, in fact, down through wood, in the correct path.
		Now, when we do that as humans, we are using three complementary sensors to measure the position of the drill and to calculate how much farther the drill has to go.
		We measure the position of our arms with sensors in the muscles and parallel with the muscles and in series with the tendons, and we get proprioception.
		That way, with visual confirmation of position, a mechanical system, a robot, it is very easy to measure the angles of the joint and calculate position in that way.
		It is also easy to measure the load that is applied by the end effector.  For a robotic surgical system, some sort of a fiduciary marker, the analog to visual location is needed so that we can figure out where the effector is in relation to tissue that is being operated on.
		Hand operation of controllers, of course, is easiest with, for motorized stereotaxy, where the device is simply under the control of the operator, visual feedback is satisfactory.
		There have been some applications of virtual reality endoscopy for cholecystectomy where there is also continuous feedback, and the operator is at all times visually, either directly or through a camera, seeing where his end effector is going.
		A second application that is being proposed for robotic surgery is telesurgery where the issue of delayed visual feedback between the effector and the operator becomes an issue.
		We will be talking today about programmed operation where the operator really does not see where the hole is being drilled.  He has some indirect sensors of where the hole is, but basically, the robot now at this highest level, has taken over control and is going ahead and advancing the effector through the workpiece up until the point that the hole is completely drilled.
		The effectors, and we can go through that fairly briefly, are either passive or active.  The one we will be talking about today is an active effector, the end mill for femoral reaming for cholecystectomy, scissors and sutures, general surgery.
		The real issue for the engineer is to look at the potential error sources.
		A surgeon with a tremor obviously not going to be able to do as effective a job and stability in closed loop feedback systems is a trade-off in measurement between how quickly I can move my workpiece and my effector to where I want it to be, and how quickly I can measure where I have been.
		A classic example of this is a thermostat that reacts too quickly to a change in heat and suddenly, as you open the door, the thermostat thinks it is freezing cold and heats the room up to 80 degrees before the thermostat has had a chance to recognize that this is not a long-term change in temperature, but merely somebody opening the door and leaving it open for a few seconds.
		In any engineering system for closed-loop control, there is going to be a trade-off between over-shoot, reacting too quickly, and slow response, which we have talked about.
		An even more serious control besides tremor, is the loss of proprioception in a robotic surgery system.
		As soon as a sensor no longer knows where the work-piece is or where the effector is or where is that drill bit.  As soon as we lose control of the knowledge of where that drill bit is, then we no longer know where we are drilling a hole.
		Then there needs to be some sort of a fail-safe mechanism that will either stop the drilling, retract the work-piece, or have some sort of a redundant sensor that says that if one sensor doesn't know where the hole is, another sensor does.
		Several fail-safe modes need to be incorporated.  Three of those that are commonly used are a watchdog at the start-up to make sure that the system is operating properly, if a failure occurs to freeze the drilling in the last known position, and some sort of a retraction to a safe park-zone.
		All of those are significant engineering issues that we need to talk about in response to that first panel question.
		That is it for the initial presentation.  Dr. Hannaford, I think, will do one specific to the closed panel meeting.
		DR. YASZEMSKI:  Thank you very much, Dr. Walker.
		I would like to now ask for Dr. Laurencin to present the clinical review.
		DR. LAURENCIN:  Thank you.
		Mr. Dillard, Mr. Melkerson, Mr. Demian, thank you for inviting me to present some of my views on robotics.
		The proposed use of robotics in surgery is every increasing.  A number of scientific and clinical developments are probably responsible for this trend.
		First, the current emphasis on minimally invasive surgery, for example laproscopic procedures as just talked about, have brought great interest in robotics.
		Robotics offers distinct advantages to stability and the ability to work with precision at small scales, and again, it is well-suited to minimally-invasive needs.
		Second, the emergence of sophisticated three-dimensional patient data, usually by CT or MRI, and the software to manipulate these data, have driven increased interest and use.
		Third, for orthopedic surgery, the successful addressing of previous weak link problems such as implant materials and implant design in the 1970s and 1980s have lead scientists to new areas, in other words, robotics, to improve existing surgical procedures.
		In general, just as a bit of review implantation of robotically- assisted procedure involves planning, registration and navigation steps.
		Implanting images are taken of the region of interest and are presented to the clinician in meaningful form.
		For registration, a correlation of the image data is made with the patient's anatomy.  This is often done with fiducials or markers, as in the first generation robotic system for a hip replacement.
		Other techniques include optical tracking techniques that can obviate fiducials.  That track, for instance the curvature or surface of object and correlate them with image data.
		Finally, with data correlated to the patient, navigation or guidance can take place.  This can be performed by the physician alone, the robot alone, or somewhere in between.
		The level of involvement of physician versus robot depends upon issues of safety, physician comfort and acceptance, practicality of implantation, also cost.
		What is the particular attraction to orthopedic surgery?
		Bone, as a tissue, is relatively facile to manipulate in comparison to soft tissues, and the level of deformity in cutting is relatively low.
		Thus, on a theoretical basis, one can envision developing surgical procedures where pre-operative plans can correlate with the robotic procedure.
		In orthopedic surgery, the system that has received the most attention is the ROBODOC system, a brainchild of the early 1980s.  It goal is to ream the femoral canal more precisely in order to decrease the short term complication of femoral fractures which can occur as part of the reaming, broaching, press/fit implantation procedures of total hip replacement.
		There is also a strong suggestion that with proper fit and fill from the literature, that long term outcomes will be improved.
		Also, in orthopedic surgery, the Hip-Nav system has received attention.
		This guidance system for acetabular cup placement is designed to optimize cup position to minimize the chance of impingement, with subsequent dislocation occurring.
		A registered pelvis is used in conjunction with the tracking software to provide simulations of range of motion with cup position.
		Significant interest is present in robotic use in conjunction with total knee arthroplasty.  The robotics range from surgical assistants that hold the knee to those that determine knee alignment and perform bone cutting.
		Happy mediums have been designed between robot and human implements.
		For instance, Davis, et al. have designed a cutting system operated by robotic control, but hand-guided by the surgeon.
		In effect, a virtual jig is formed, placing force to keep the surgeon's hands on track when making bone cuts.
		Areas of the spine may, in many ways, be some of the most challenging of all work applications of orthopedic robotic technology.
		The precise placement of pedicle screws present special problems in registration and demands for absolute precision in application.
		The stakes are large.  A robotic system that could be used with confidence might allow the routine use in pedicle screw placement.
	Where is the technology going?  Clearly with advancements in the areas of control, sensor design and mechanical engineering which we just talked about, the capabilities of these robotic devices will dramatically and we should look forward to their increasing role in surgery.
		The better question now is where is the technology today?
		Today's technology does have limitations, from industrial-based mechanical manipulators with only first or second generation optimization, to work in clinical environments, to issues of sterility maintenance.
		It should be remembered that the same technology that drives robotics is the same technology creating our Y2K anxieties; in other words, the computer.
		No system can guarantee complete safety in all settings.
		Inventors have addressed this fact by mechanical design changes such as using the low-pressure pneumatic manipulators, or by placing more control in the hands of surgeons.
		But that fact still remains.
		Physician comfort and acceptability is growing via focus groups, the media and scientific presentations.
		However, essential questions of how this technology affects my patients outcomes and what are the costs involved, must have no-nonsense answers for the technology to find wide-spread and lasting clinical use.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		Let's go around the table now, and we will begin with Dr. Silkaitis, and ask each panel member to comment or to ask for clarification from the FDA of informational or procedural points.
		Dr. Silkaitis?
		DR. SILKAITIS:  Yes.  The area of robotics is certainly a large area for consideration.
		Are we looking at robotics in a specific area of its use?  In other words, are we talking about active robotics?  Are we talking about semi-active robotics?
		DR. YASZEMSKI:  Who are you directing the question to, Dr. Silkaitis?
		DR. SILKAITIS:  To FDA.
		DR. YASZEMSKI:  Mr. Dillard, would you care to comment on that?
		MR. DILLARD:  Yes.  Jim Dillard.
		I think the focus here today, or where we would like you to focus your attention predominantly, is on the increasing use of this technology.
		I think that both Dr. Laurencin and Dr. Walker gave us some examples from the simplistic or the types of technologies that are used today that are not computer-controlled, or minimally computer-controlled, all the way up to those which are under great computer control as well has having a lot of mechanical interactions with them that are under computer control.
		Our main concern is regarding that end of the spectrum of the technology.
		We are seeing ever-increasing amounts of devices that we are faced with making either regulatory decisions on from the stand point of should they be on the market or not, as well as the design of clinical studies in order to evaluate these kinds of technologies.
		The main focus we would like you to have today is on this area of technology in orthopedics, number one.
		There are other indications in usage in other areas of medicine, but we would like you to focus on orthopedics.
		We would also like you to focus on those types of issues, and this is really more of an issues-based discussion I think, from either the engineering and/or the clinical perspective, what are some of the questions that we should be asking.  What should FDA be asking?
		Any guidance that you might have into types and ways to evaluate the technology and the important things to look for will help us.
		I think we are struggling right now in a lot of areas, trying to design the right clinical studies, as well as what is the right amount of pre-clinical information that we need in order to evaluate the technology as well as move on to clinical studies.
		I hope that helps.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Silkaitis, with that, what are your thoughts as the perspective of the industry representative regarding either the engineering or the clinical issues?
		DR. SILKAITIS:  In other words, the evaluation or the discussion is centered on the equipment that is being used to perform the surgery as opposed to the evaluation of a device where we are looking at longer-term data.
		The question is what is the least burdensome amount of data that is necessary to demonstrate that the equipment meets its performance characteristics.
		So, in a particular case, if the robotic is to drill a hole or drill a cylinder by certain dimension, then we take a look at, and we can easily measure, how precisely it does that, how often it does it, and what the user errors are involved in achieving that.
		So, I guess from my perspective is that we are not looking at device designs, but we are looking at equipment performing to its characteristics.
		DR. YASZEMSKI:  Thank you, Dr. Silkaitis.
		Mr. Dillard, we are coming around.  I will give you the opportunity to add again.
		MR. DILLARD:  I think, at this point, that is what I would have added without Dr. Silkaitis' question, so with that I think I will pass.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Skinner?
		DR. SKINNER:  Well, I'm certainly not a robotics expert, although I am an orthopedic surgeon so I guess that makes me into something of a robot.
		I think the issues are the same ones that Dr. Silkaitis mentioned.
		If the issue is number one, that we are going to cut a hole or cut a surface, then I think it is a matter of accuracy how closely we come to where we want to cut that hole or surface, and it is a matter of precision as to how precisely we do it each time.
		And it is a matter in comparison to what a surgeon can do.
		I know Dr. Bargar has compared himself to a robot at times, and has turned out nearly as good as a robot.  Maybe he will comment on that at some point.
		But, if we are going to get that accuracy and precision that is better than a surgeon, then I think that has to be the criteria.
		If it were only doing as well as a surgeon, then we have a cost issue to deal with.
		On the other end of the things, when it comes to comparing this surface we have cut or drilled or whatever with a robot and we compare the clinical results, again I think we have to consider that this is simply a surgical tool, and the immediate clinical results are the problem we have to deal with.  That's the issue.
		It is not what the results are six weeks later, six months later, six years later.
		It is a surgical tool and what happened, basically, the day after surgery when you look at the x-rays or whatever criteria you are going to look at.
		DR. YASZEMSKI:  Thank you, Dr. Skinnner.
		Dr. Larntz?
		DR. LARNTZ:    Well, I'm a statistician.  I think I said that this morning.  I will say it again, just to make sure we are clear.
		I am not a robot, I think.
		But I do have some appreciation for computer technology and I have some concerns about computer technology, as a long-time programmer.
		I guess I would say there is no such thing as bug-free software.  If someone claims that then you have just got someone who is a liar.
		So, I think we have to very carefully consider that the technology will do what it is supposed to do.  Computer-assisted technology.  It is very difficult.
		Systems get upgraded, and isn't it amazing that every time there is an upgrade what happens to your system?  Any ideas?  You have all gone through it.
		Things don't work as well, so you have to be very, very careful.
		Now, given that caveat, I am incredibly in favor of developing technology-based assistance because why?  There is going to be consistency.  Whatever this thing does, it does it consistently.
		We found in lots of areas, consistency is a very good thing, in and of itself, once you understand what the result is.  That consistency has incredible value.
		Look at the cars you drive now compared to what they were 20 years ago.  Consistency is incredibly important as a result of the quality movement.
		So, I think there is an advantage there that could be beneficial in lots of ways.
		One of my cardiologist friends would say you might avoid what we call operator error.  And operator errors do occur on occasion in surgeries.  And sometimes, if something is being done consistently, it will avoid that.
		From the statistical point of view, how should we evaluate things?
		We should evaluate these technologies in the same way that we evaluate every other new device, every other new item that we are doing.
		Does the technology do what it is supposed to do?  That is first.
		And then, what is the benefit of that?  Is the benefit to the surgeon?  Is the benefit to the patient?  Is the benefit to, well, whomever.
		Now, what kind of benefit can you expect?  I think there is a whole range of things we have heard people say.  There might be short-term benefits.  There might be long-term changes and benefits, too.
		We have to be very, very careful in evaluating what those would be.
		Cost savings or cost increases, sounds like it could go both ways.  And there's lots of various things.
		So, my opinion, short bottom line is we should evaluate these new technologies in the same way we have always evaluated technologies		.
		In fact, we should always evaluate things better than we have been doing, which is to say we should use well-designed studies and carry them out carefully and not just say gee-whiz, wow, this works!  Let's do something with this.
		Be careful, think about it and use the same principles to evaluate these as you would use any other medical device or technology.
		DR. YASZEMSKI:  Thank you, Dr. Larntz.
		Dr. Laurencin, you presented your review, but I would like to offer you an opportunity at this time to add additional comments that you might have.
		DR. LAURENCIN:  I'll be making comments later this afternoon.
		Thank you.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		Dr. Besser?
		DR. BESSER:  Yes, thank you.  In answer to I guess the first -- from the engineering perspective, the mechanical engineer, and I never thought orthopedic surgeons were robots; I knew they were carpenters.
		So, I very much enjoyed Dr. Walker's reference to drilling holes for pegs for a coat rack.
		I think some of the important issues that the FDA has to be aware of, and some of the I guess assumptions that we have implicitly made, may or may not be true.
		First, if there is an ability to visualize directly in real time, so that you can see if your device is not drilling the hole in the right place or at the right angle or is not putting in nails or screws or whatever it is that you would like to do orthopedically, so that you can stop it, then those types of systems you are talking about are essentially remote-manipulator Waldo kind of system.
		I think that issues inherent with systems like those are those in any mechanical linkage where you want to look at backlash in the gearing system or the ability to precisely position a device and know that it is going to stay there and be rigidly there and not going to move as you start to use that tool against whatever surface you are working on.
		That merely by applying pressure to it, your linkage isn't going to deflect or deform or in some way not drill the hole where you want it to be or nail the nail where you want it to be.
		If you are not able to visualize your end effector and now you are flying blind, then you need a way, before you start to cut, drill or screw, to know that you have positioned that end effector to the accuracy necessary.
		You have to have some way to address that, either through some imaging system that is used after you have positioned your end effector and are ready to cut, drill or whatever.  I would think that you would have to be able to ensure that you have positioned the end effector appropriately.
		Possibly, after demonstrating your ability to position that end effector appropriately, 100 times out of 100, then you can allow them to continue using this without that first visualization.
		I guess I am looking at the safety aspect of this as opposed to the effectiveness aspect.
		My first concern would be is this safe?  Is it going to cut, drill, screw in the wrong place?
		Then, looking at effectiveness, one issue I would sort of like to throw out to the orthopedic surgeons is we are sort making an assumption that this clinical end point is extremely dependent upon your ability to precisely and accurately cut, drill, screw.
		We should not require these systems to be more accurate and more precise than is currently being done by a surgeon.
		If a surgeon can perform some operation and get good clinical outcomes and is not able to cut to 1/1000 mm or a tenth of a degree, then there is no reason to try to build a machine that can do that.
		An orthopedic resident friend of mind once told me that the perfect is the enemy of the good.
		When you are an orthopedic surgeon and you keep working at it, trying to get it absolutely perfect is usually when everything goes south.
		So, I think that when evaluating systems like this, I am not sure that we should hold them to the standard that they have to be better or more accurate than the skilled orthopedic surgeon.
		If you can do it as well as the skilled orthopedic surgeon and in less time so that the patient has to endure less time in surgery or making it easier for the orthopedic surgeon to do what he is trained to do, then I think that is a valuable end point and a valuable goal for this, also.
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		Dr. Cheng?
		DR. CHENG:  Well, I guess I am of the opinion that surgery is really done in your head, not with your hands.
		But, these technology-assisted devices can be valuable in performing mechanical tasks in the operating room.
		So, I guess if the FDA wants to evaluate these, I would encourage the FDA to determine what is the goal of the device?  If it is a scalpel, a laser, a coagulator, or whatever it is, does it do what it is meant to do?
		Secondly, as a result of that, if it meets that goal is it actually meaningful from a clinical standpoint or is it meaningful from a financial standpoint.
		I think that orthopedic surgeons have a very low threshold, historically, for embracing new technology.  However, we have to back away and ask ourselves does this actually makes sense to use this particular device.
		That would be the limit of my comments.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Dr. Hannaford?
		DR. HANNAFORD:  I want to start with a couple of words about industrial robots because many of the initial laboratory systems are based on industrial robots, and some of the commercial systems are based on industrial robots, and a lot of the thinking about robots is based on industrial robots.
		Industrial robots have a couple of attributes that are driven by their existing markets.  An amazing percentage is simply that one task of spot-welding that we see in the auto commercials all the time.  That is a huge bulk of all the robots in the market, but not all of them.
		But the two things that they are sold on and deliver on are accuracy, or more precisely, precision, and reliability.  Over the years they have a track record of doing very well on those things.
		The reliability requirements of manufacturing are so much greater than any conceivable volume of surgery we could ever see being done, that a robot in manufacturing will have to do that task precisely and stay within specs thousands of times per day and operate for a year or more.
		So, in the sense of how long will it last and stay in its performance range, that is a much, much more demanding realm than surgery.
		Now, let's look at safety.  Safety is a huge concern in manufacturing as well because the manufacturer is liable, and so forth.
		But the traditional safety approach in industry and manufacturing is that you put a cage around the robot and you keep out of reach of the robot.
		That is a very effective approach, but it has no usefulness for robotic surgery.
		So, we do have to think very, very carefully about safety, even though this base technology that is coming into the OR is already very reliable.
		In surgery we have to adopt a totally different safety approach than is used in manufacturing.
		In particular we want to look at the control system.  This has come up in the software comment that we heard.
		The control systems contribute to the reliability of industrial robots, but are sometimes, to a greater or lesser extent in the different systems I have seen, modified for surgical applications.
		So, that is where engineering attention, and this now gets me directly to the first question of design review and so forth, should really be focused.
		The typical approaches are adding extra sensors, redundant sensors, so if a sensor fails that state can be detected right away.  Sometimes, such a modification is really an add-on to the intact control system.
		But other times the system is connected to something else in such a way that its properties belong to a bigger system and that has to be looked at carefully.
		Some of these systems have a form of surgical assistance where force information comes back to the surgeon through a control device.  This is known as a force feedback system or a bilateral system.
		That system has to be carefully analyzed as a whole and not just certified based on the safety of all the individual components.
		Finally, the last point I want to make is training.
		I think the type of and nature of training of surgeons who will use these systems is very important.  It is precisely the reliability of the base technology that I think makes it very important.
		I am worried about a hypothetical situation where a system may have a big red emergency stop button for a surgeon to use in case of some problem, but it may work so well for a couple of thousand procedures, that if something does go wrong on the next procedure, the physician may not remember where that E-stop button is.
		On the other hand I think that these robotic systems present an opportunity for safety because I think they lend themselves, in many ways, to training through simulation which people are working on for conventional surgery, but is very hard.
		In some cases it is easier to do that kind of training with a robotic system.
		I think about the example of flight training for pilot, where a pilot will have to practice a situation that is very, very rare, such as having an engine fail during take-off.  How many of us have had that happen when we have been on a plane?  Very few of us.
		Probably doesn't happen to most pilots in their whole career.  Yet, all of them are trained to do something about it in simulators.
		I hope that robotic surgical systems will include that kind of training.
		That kind of training, if it is done in simulation, can be done periodically, and a surgeon can be recertified.  So, that surgeons are actually ready when some very, very rare problem comes up.
		I really don't view that as a problem with the technology as much as an opportunity to do even better.
		So, those are my comments at this point.
		DR. YASZEMSKI:  Thank you, Dr. Hannaford.
		Dr. Aboulafia?
		DR. ABOULAFIA:  I don't have any specific comments right now.
		DR. YASZEMSKI:  Thank you.
		We have gone around the table now, and I would like to end by asking the two people who were the lead reviewers to close up with any additional comments they have come up with after listening to the discussion.
		I would like to start with Dr. Walker.  Any thoughts after the round table discussion?
		DR. WALKER:  Well, I want to thank Dr. Hannaford for adding some additional safety considerations that I didn't include in mine.
		I think that as we go around the table and consider safety, the points he raised about training, as well, are extraordinarily important.
		The argument of this device providing greater precision, but the safety issues and the down-sides are what we need to be worried about in the regulatory environment.
		DR. YASZEMSKI:  Thank you, Dr. Walker.
		Dr. Laurencin, any additional comments?
		Before we go, let's ask Dr. Skinner.
		DR. SKINNER:  Yes.  I want to thank Dr. Hannaford for his comments, too.
		I wanted to comment on his comment about the robots that do the car welds.  In those situations, it is a significantly different situation because I think the car is in the same spot each time and the car is the same size each time, and the robot knows exactly where it is going to go each time.
		In a surgical thing with a robot, and I have only done this once with H.A.P. Paul about ten years ago, the cutting location for the robot is determined by the surgeon, pre-op, and there can't be any screw-ups with the software or the surgeon doing that.
		The cutting location is determined by the surgeon in registration at the time of surgery so that the robot knows where the bone is.  The bone location is determined by the surgeon, and hopefully not moved during the cutting process.
		So, there are multiple potential areas for problems to occur where there shouldn't be any problems that occur.
		I think these are the issues that make it different from an industrial robot.
		While I say that, I don't want anybody to think I am against robots.  I think that it is something that is going to come and it will be very helpful to surgery in general, and probably orthopedics in particular.  I don't know when, though.
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Dr. Laurencin?
		DR. LAURENCIN:  I'll just close in that traditionally orthopedic surgeons have always embraced new technology.  Ninety per cent of the operations that we do involve new technology.
		If you look at surgery such as arthroscopy, thirty years ago it really didn't exist.  Total joint replacements.  Everything in our generation, a generation ago, really didn't exist.
		So, we traditionally embrace new technology.
		One of the issues that comes up, in terms of this group, is what sort of endpoints we should be looking at short-term and long-term to determine whether this new technology will be viable or not.
		My feeling is that it may take, at least for the first new materials coming through, an over-evaluation in terms of endpoints just to make sure that all bases are covered.
		For instance, if we are looking at a total joint replacement we may have to look at the endpoints that we traditionally look for even in a new device for total joint replacements.
		You may say that this is very different from a new device.  But in many ways, the types of procedures that are done in terms of some of the more advanced procedures that are done with robotics are actually are creating a new way that a prosthesis may function.
		So, we have to consider, in terms of what our endpoints are, we have to start by I think looking at the endpoints that we traditionally use for total joint replacements and then say are these appropriate endpoints for this sort of situation.
		Understanding that in the first couple of ones that go through, we may be looking with a fine-toothed comb.  But that over time, when more are accepted, we will have a basis for moving down from there.
		DR. YASZEMSKI:  Thank you, Dr. Laurencin.
		DR. HANNAFORD:  Mr. Chairman, could I just briefly respond to Dr. Skinner?
		DR. YASZEMSKI:  Dr. Hannaford.
		DR. HANNAFORD:  Thank you.  Yes, Blake Hanford.
		I very much agree with your comment.  So, I just want to clarify that all my praise of industrial robots was not meant to say that they are automatically safe in this kind of context, by any means.
		I was really referring to the robot arm as a component in this kind of system.
		So, my remarks about the fact that these control systems are modified and expanded into bigger systems, address your concern.
		So, I thank you for clarifying that.
		DR. YASZEMSKI:  Thank you, Dr. Hannaford.
		At this point I would like to give a short summary of the discussion from Mr. Dillard and the FDA, then proceed to ask whether we've answered the questions they posed to us.
		With respect to the first question, the issues and engineering concerns, the panel felt that a main issue was does the equipment meet its performance specifications.
		That is, if the issue is to cut a hole or cut a surface, then we would suggest to the FDA that the necessary data, from an engineering perspective, is did the equipment do that and do that safely.
		Safety issues came up repeatedly through the discussion.  The general feeling was that accuracy and predictability of the cut is paramount and that safety should be the number one issue.
		So, if blind positioning, especially of the effector occurs, then there must be some sort of registration, be it determined by the surgeon directly by anatomic means or by some surrogate means, to be sure that the effector tip is where it should be prior to beginning its cut.
		With respect to question number two, clinical study endpoints, we heard from Dr. Larntz that we should use the same type of controlled studies that we would use to evaluate total joint arthroplasty, in either the short or the long term, without the robot.
		And really try not to deviate from that, and to be certain that we don't make the statement that we have a new tool and all we have to look at is the tool.
		Dr. Laurencin reminded us that we have well established clinical endpoints for total joint arthroplasty outcomes that we have used over a variety of generations of equipment, and that perhaps we should continue to use those that are tried and true.
		I think those would be our thoughts for both questions two and question four.
		With regard to question three, Dr. Skinner made the comment that the day after surgery should be the time from short-term to assess whether if any sentinel events occurred, perhaps new complications that don't currently occur with surgeons who are doing this manually.
		We should look in the short-term for new things, specifically large complication type things.
		I think this summarizes our thoughts on this and would ask Mr. Dillard if we have answered these questions to the FDA's satisfaction.
		MR. DILLARD:  Jim Dillard.
		Thank you, Dr. Yaszemski.
		I might have one follow-on that perhaps ties a couple of these together.  I appreciate everyone's comments because I think they will be very helpful.
		One of the things that perhaps we are struggling with the most, and I think you were pretty clear in some of your comments about clinical endpoints and utilizing the clinical endpoints, certainly in the early term with these types of technologies, and that will be very important.
		One of the questions that we get repeatedly in this area is, as I would term it, perhaps, a tool approach versus the clinical outcome approach.
		I think that many companies in this area are quite concerned about being judged to a standard that might be a higher or at least as high as a standard for the new particular implants that we have, that Dr. Laurencin talked about, versus what their product is specifically intended to do, which is to be a tool that cuts, shapes, mills, reams, et cetera.
		One of the greatest struggles I think we have is the issue between surrogate and clinical endpoints.
		I just was curious whether or not anybody had any comments about how to tie those two together, and if there were any circumstances where one might see that surrogate endpoints might be adequate enough, or under all circumstances would this panel recommend that clinical endpoints is where the focus ought to be, from the standpoint of the FDA?
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Mr. Dillard, before I open that to the panel, may I ask for a clarification?
		Have there been any suggested surrogates to the FDA that we might consider specifically at this point?
		MR. DILLARD:  At the risk of not being able to disclose too much information with the companies that may have products under review, I think the concept might be when Dr. Skinner was talking about the accuracy with which cuts can be made, and the ability to be able to determine how accurate and reproducible those cuts might be, then is it adequate enough to look, in the short-term	, at the performance of those cuts and whether or not we have good short-term outcomes, based on what the product is intended to do, versus an effect on the long-term outcomes and clinical performance when you actually the place the implant and you look then at the surgery plus the implant and what the effects may be.
		How do you tease those out?  The effects due to the tool and the effects due to the implant.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		I would like to open that up to the panel.
		Dr. Aboulafia?
		DR. ABOULAFIA:  I can't give you an example about a specific product for reasons of protecting industry, but I think it depends.
		To use your example though it may not be the best, if you are looking at the accuracy of the cut, and you are making the cut the same way you do all the other times, then it probably isn't important to get any more than 24 hours long-term follow-up.
		But if you are measuring that cut and then using an instrument other than what you normally use to cut the bone, then maybe it will be a difference.
		Maybe there is heat generated from the device which is different than standard transverse oscillating saw.
		Maybe the heat generated from making that cut may have an adverse effect on the fixation to the bone, and in six months you might see mechanical loosening and complications.
		So, I don't think that you can say,  categorically, that everything is going to be the same.  We need long-term follow-up on all of them.
		I think you do have to tease them up to say is it really just measuring an ankle and everything else is the same or are we doing something inherently different with this after we do that.
		DR. YASZEMSKI:  Thank you, Dr. Aboulafia.
		Dr. Cheng?
		DR. CHENG:  I think I would just mention to the FDA, it depends on what the manufacturer or the sponsor is claiming the device does.
		If it actually improves the patient outcome, then they have to show that.
		I have no doubt that a machine can precisely do some mechanical act better than I can do it.  There is no question.
		But the second part to my initial comments, is the result meaningful really begs a question.
		Is it useful in surgery, number one?
		How is it useful?  Does it make the patient's outcome better in some way?  Does it make the ability of the surgeon to implant something or do some particular task better?
		So, your questions are a little bit vague and hard to answer because they aren't specific enough.
		But in general, I guess we would go back to what the sponsors are claiming the device will do.
		DR. YASZEMSKI:  Thank you, Dr. Cheng.
		Other comments? Dr. Besser?
		DR. BESSER:  I guess a question to the FDA.
		When a company brings an orthopedic implant to the FDA for approval, usually along with that system is instrumentation for making the cuts that are required to implant that.
		How are those instruments currently evaluated?  Are they evaluated separately from the device or merely as a clinical endpoint after you have the whole device in?
		DR. YASZEMSKI:  Mr. Dillard?
		MR. DILLARD:  Thank you.
		I would say that perhaps there are two different circumstances.
		One would be dedicated surgical instruments that come as part of the kit with the implant or as a stand-alone basket of tools that go along with a line of implants.
		Many times those particular tools are not looked at.  Manual surgical instruments, for example which is a category of products, that you can find in our code of federal regulations, are currently exempt products from pre-market notification.
		If there are special kinds of manual surgical instruments or special kinds of orthopedic surgical instruments, that are very specific to a type of procedure, for example, which come with a new indication for use or a very new kind of technology, many times we will evaluate those at the same time that we are evaluating the new type of technology.
		So, I think, as the statement goes, there is not an easy answer.
		There is probably the spectrum here of those products which are very much like manual surgical instruments, other orthopedic instruments that are commonly used across many procedures that would be exempt from pre-market review, to those that are very specialized, may come with their own intended use, and are for a specific new type of technology that would be evaluated with that technology.
		DR. BESSER:  Dr. Besser.  If I may follow-up.
		Then for that second category, where it comes for a very specific use such as for an orthopedic implant, are there surrogate endpoints that you look at for that positioning jig or cutting device, or are the only endpoints that you are looking at the clinical endpoints, long-term, was this successful surgery?
		DR. YASZEMSKI:  Mr. Dillard.
		MR. DILLARD:  I think in that kind of situation, many times if it is a new implant that needs clinical study, for the new implant we will also look at the type of human factors, we will look at evaluation of the types of tools that go along with the implant.
		We tend to take more of a procedural look.  Is the procedure, which includes the physician, the implant tools, as well as the new implant, what is the overall success of the total procedure?
		We tend to label it from that particular vantage point.
		That isn't to say though, that there aren't specific tools that are manufactured to do a procedure that don't include a prosthesis, for example, that might not have their own evaluation.
		Sometimes, they do if it is a new type of tool.
		Many times, if the questions are in the short-term and what the effect is, we tend to focus on those issues that need to be answered for that particular type of tool.
		So, I think we have tried to take the approach of what are the appropriate questions and issues that need to be addressed for that particular type of situation, and tried to focus our attention on that.
		So, I am echoing a little bit of the vagueness of my answer to Dr. Cheng, to try to get you all to address both of those kinds of situations, in this particular case.
		DR. YASZEMSKI:  Thank you, Mr. Dillard.
		Dr. Besser?
		DR. BESSER:  Then, I would say, in response to the FDA question that, yes, it depends on what we are looking at.
		If you have a device whose specific task is to cut a line at this angle, then I would say your first surrogate endpoint is absolutely, did it cut that line at that angle?
		But then you can't do the operation was a success; the patient died kind of thing where you also want to look at the clinical endpoint.  Now that it has cut that line at that angle, did that help?
		DR. YASZEMSKI:  Thank you, Dr. Besser.
		Dr. Skinner?
		DR. SKINNER:  I want to comment on Dr. Aboulafia's comment which I think was very worthwhile and very important.
		Such things as thermal damage could cause a change in the bone in a femoral canal, for instance, over a period of time.
		I think though, that it would be more likely, since the prosthesis fits the bone perfectly after one of these robots cuts the hole for the prosthesis, that what we are looking at is a situation where the bone and the prosthesis fit together perfectly the day it is put in, and the next day the bone starts remodeling.
		So, any changes you see six weeks or six months or six years later might be due to the prosthesis and not necessarily the cutting.
		I think that adds a variable that makes the interpretation much more difficult.
		I would still lean towards considering it to be a tool.
		DR. YASZEMSKI:  Thank you, Dr. Skinner.
		Mr. Dillard, may I submit to you and the FDA that the additional discussion has led us to  comment that there is a great deal of uncertainty, at this time, to your question, and that it appears that surrogate endpoints seem to be appropriate, but we would like to suggest to the FDA to reserve caution.
		To not rely only upon them, but also give consideration to other long-term changes that may arise as a result of this new technology.
		Have we adequately answered the FDA's questions at this point?
		MR. DILLARD:  Yes, I think you have provided us with good guidance, and I appreciate the discussion.
		DR. YASZEMSKI:  Thank you, very much, Mr. Dillard.
		We will now proceed to the closed session.
		I would ask that we clear the room because the remainder of this meeting is closed to the public.
		We will take a five minute break while the room is being cleared, and only previously designated individuals, who have proper identification, will be permitted to stay for the closed session, scheduled to discuss a clinical study.
		Thank you, very much.
		(Whereupon, the proceedings went off the record and then resumed in Closed Session.)














??



 

 




	NEAL R. GROSS
	COURT REPORTERS AND TRANSCRIBERS
	1323 RHODE ISLAND AVE., N.W.
(202) 234-4433	WASHINGTON, D.C.  20005-3701	www.nealrgross.com