“Is patient specific instrumentation (PSI) less expensive? Is it easier for you? The answer is definitely ‘no, ’” argues Paul Lachiewicz. “PSI provides a way to position the implants specific to that patient’s anatomy (preop valgus, patient height, hip pathology, etc.), ” says Mike Berend. And you can do it before you enter the OR…with templating this means it is time neutral.”
This week’s Orthopaedic Crossfire® debate is “Patient Specific Instruments: Overpromised, Under Delivered.” For the proposition is Paul Lachiewicz, M.D. from Duke University Medical Center in Durham, North Carolina; against the proposition is Michael E. Berend, M.D. from the Center for Hip and Knee Surgery in Mooresville, Indiana. Moderating is Cecil H. Rorabeck, M.D., F.R.C.S.(C) from the University of Western Ontario.
Dr. Lachiewicz: “In their brief lifespan these instruments have overpromised and under delivered. There are now five bad ideas in total knee arthroplasty (TKA): carbon-reinforced polyethylene, metal-backed patellae, low-molecular weight heparin, and gender-specific knees…and now, patient-specific instruments (PSI). This is a reincarnation of computer-assisted surgery that shifts the work from intraoperative to preoperative. You must do a CT or MRI, the surgeon has to do a lot on both the input and after the technician returns the plan. You have to adjust the plan for deformity, contractures, and ligament imbalance.”
“The marketed advantages of PSI are the same ones that were touted 15 years ago for CAOS [computer assisted orthopaedic surgery]. There was never any proof that avoiding IM instruments affected blood loss, emboli, or cognition. The claim that they avoid outliers has not been proven. The claim that they get better outcomes and deliver better patient satisfaction has not been proven. These instruments are even marketed as being less stressful for the surgeon…are you kidding me?”
“The disadvantages? The MRI is going to cost approximately $1, 000 and the jigs are $1, 500. You have to take into account the time it takes your office staff to schedule the MRI, the learning curve, and the time it takes you to review and adjust these plans. Is it $500/hour? Are you getting $500 more when you do these? Certainly not.”
“In North Carolina, one insurance company is fighting back when surgeons are ordering MRIs. Who will pay for this? Several studies show that these will work, but the vast majority in North America and Europe show that they don’t work. Stronach and Peters et al. (CORR [Clinical Orthopaedics and Related Research], 2013) have done a study where he had to change the size of the vast majority of femurs and tibias (roughly 2.5 per knee). There was no difference in tourniquet time, and he cautioned against using PSI plans.”
“A cost effectiveness study by Slover et al. (Journal of Arthroplasty, 2012) looked at the Markov decision model and found that the PSI plans were much more expensive. They calculated that you would have to reduce your revision rate by 50% to be cost effective—not going to happen.”
“Do PSIs improve alignment? Robert Barrack and Ryan Nunley have done multiple studies on this, one of which is a three cohort study (one conventional versus two PSI systems) published in CORR in 2012. One of the PSI systems is no longer available. They did postop CTs and found that the number of outliers was the same in all groups; there was perhaps even more valgus outliers with both PSI systems.”
“Are they cost effective? The thinking is that there will be fewer trays and that it will save the hospital money. In this study, they spent 12 minutes less in the OR, there was no difference in alignment, and they saved the hospital $25 per case because they processed four fewer trays.”
“PSIs are not going to balance the knee or do ligament releases; neither will they tell you the proper tibial component rotation. They are not going to resect the patella and they’re not going to cement the components. And it’s these steps that determine the long term durability of TKA!”
“In a study I conducted with Drs. Del Gaizo and Soileau (Journal of Knee Surgery, 2009) we found that you should template your total knees for both primary and revisions. Template the sizes and the amount of resection…and have a plan going in. Brett Levine has taken this one step further. He has done digital templating and supplied the sizes to three vendors and told them to reduce the number of trays. In 97% of his cases the prepared sizes and three trays (prepared by the implant manufacturer) were all that were needed.”
“When you consider any new technology for TKA, ask yourself: ‘Does it provide better outcomes and decreased revisions for your patients? Is it less expensive? Is it easier for you? For PSI the answer is definitely ‘NO.’ Put the onus on the implant companies. They should do something to decrease their instrument trays and weight.
Dr. Berend: “Obviously, my friend Paul is stressed even thinking about PSI, thus it’s probably not the best tool for his OR. We’ve studied this for 15 years and found that tibial component varus greater than three degrees in patients with a BMI over 33 meant over a 100-fold increase in failure (Berend et al., CORR, 2004). So the conclusion of alignment as an independent variable not affecting outcome may be valid. But when you combine it with other patient and implant factors it is critical.”
“If you look at the precipitous drop in survivorship in the patients with whom we missed the mark (that had a BMI over 33) there was a failure rate of 0.4% at 15 years… unacceptable. In our database of over 6, 000 total knees, for those in neutral alignment there was a 0.5% failure rate regardless of alignment. We can now say—with data to back this up—that the target we’re aiming for is between 2.5 and 7.2 degrees of valgus. All had a similar failure rate: +/- 3 degrees from the mean of 4.5. For those patients who tip into varus we found a four-fold increase in failure at 1.8%. And if you tip into valgus we’ve seen an increased failure rate go to 1.5%.”
“Regarding the mechanism of failure, on the varus side we saw tibial collapse with overload on the medial tibial bone, and on the valgus side we saw instability. If you look at the numerator, we only hit the mark in 72% of the knees. This has been backed up by many studies of conventional instrumentation. So the data indicate that alignment has a huge effect on outcomes. It’s multifactorial, with alignment, patient factors, and implant factors being important. We need cost effective, time efficient, reproducible, and transferable technology.”
“The next generation of navigated knee is preop imaging with preop planning to get six degrees of freedom to align your implants. The big message? This is an important tool that in a free society one should be able to use to plan bone cuts and the operation. And you can do this in the leisure of your own office…away from the stress of the OR.”
“So what were the promises of PSI? To provide a method or tool to position the implants specific to that patient’s anatomy (preop valgus, patient height, hip pathology, etc.). It allows you to do this before you enter the OR…and with templating this means it is time neutral. The goal of reducing instrumentation needs is an important one. Cost reduction is important, as is the improvement of alignment.”
“Even if it is 12 minutes per case, that’s important. There have been a number of studies showing that time in the patient in the OR has been reduced by 12 – 28 minutes (versus standard instrumentation)…and in a multiple joint center that is critical. (Barrack et al., JBJS, 2012; Nunley et al., CORR, 2012; Watters et al., J Surg Ortho Adv, 2011) Lowering the processing time is important if you’re paying for the staff’s salaries. And obviously there’s an additional cost with any new technology.”
“The studies show no difference in coronal alignment. They’re not worse, so if you choose to use this tool and avoid intramedullary guides then it’s important. There have been fewer outliers reported in some studies, depending on the parameters measured…the central third or three degrees from neutral. Our data suggests that anything outside of 2.5 – 7.2 may involve a higher failure rate. And perhaps for lower volume surgeons this tool may be quite helpful. But we still have to act as a surgeon and adjust things as necessary such as bone cuts or soft tissue balance. What we don’t know is if the things we adjust during surgery are important or if they have a long-term benefit.”
“Adolph Lombardi’s group has looked at custom versus conventional guides and found—head to head—a lower incidence of outliers (6% with conventional; 1.5% with PSI). In a multicenter series of 564 patients from the DeClaire Institute they found a higher percentage of people hitting the central third target (87% with PSI versus 77% with conventional).”
“PSI may be needed in routine workflow when there is extra articular deformity, periarticular hardware, or you want to avoid the IM canal.”
Moderator Rorabeck: “Paul, one minute.”
Dr. Lachiewicz: “There may be a role for this in rare cases. Mike, now I’m teaching residents to use the PAX machines to draw their cuts on the tibia and the femur…and we don’t have a lot of outliers.
Dr. Berend: “We’ve heard that from every series and if you go back and look you will see that 0.25% of your knees will be outside of 2.5 – 7.5. If you aggregate every series, conventional versus navigation, then it’s 28%. We’ve done 18, 000 total knees and we’re running at 28%…so we must improve.”
Moderator Rorabeck: “Mike, can you outline the indications for PSI?”
Dr. Berend: “I only use it in cases where I feel that my standard tools are contraindicated. We’ve used it in cases where postop confusion was an issue with the first knee…thinking that perhaps fat embolism had a role in that. We’ve used it in more severe deformities where there is metaphyseal/diaphyseal mismatch and in cases where we feel the sizing of the components is something unique to that patient’s anatomy.”
Moderator Rorabeck: “Paul, your view on that?”
Dr. Lachiewicz: “I’m not sure. Most of the surgeons in the audience can probably do these difficult knees without PSI. Cost and value are critical. I don’t think it’s been proven that patient satisfaction and longevity are improved with PSI.”
Moderator Rorabeck: “How much does it add to the cost of an operation?”
Dr. Berend: “It’s somewhere between $500 to $1, 000. I agree with Paul’s concern about who is going to pay for that.”
Dr. Lachiewicz: “What about preoperative time? Where does that get figured in? And 12 minutes saved in the OR?”
Dr. Berend: “For me, saving 12 minutes per case is to be able to do another case…that is more access to the OR, which is the major limitation to patient access in this country.”
Moderator Rorabeck: “I applaud your leadership, but at the end of the day we’re dealing with +/- three degrees. Is that going to change functional outcomes?”
Dr. Berend: “We don’t have data on that yet.”
Moderator Rorabeck: “Thank you.”
Please visit www.CCJR.com to register for the 2014 CCJR Spring Meeting, May 18 – 21 in Las Vegas, Nevada.
