A study comparing results of total knee arthroplasty (TKA) using prostheses provided by a physician-owned distributorship (POD) versus higher-cost implants provided by major commercial manufacturers found no differences in outcomes or complications.
Keith M. Baumgarten, MD, FAAOS, lead author of the study, which is published in the Dec. 1 issue of the Journal of the AAOS, said he and his colleagues decided to undertake the investigation in 2013 when they established a POD to supply Food and Drug Administration (FDA)-approved prostheses for TKA procedures in their region. “We wanted to ensure that we were providing value to both our patients and the hospitals in which we were performing TKAs,” said Dr. Baumgarten. “We defined patient value as equivalent or greater improvements in patient-determined outcomes after TKA procedures compared to the prostheses provided by the larger, Fortune 500-type, industry-controlled distributors that were already supplying knee prostheses in our region. We defined value to the hospital by providing a prosthesis that was less expensive in order to do our part in decreasing the cost of health care in our region.”
“Single-vendor contracting with orthopaedic implant companies is one method of attempting cost reduction in TKA,” the study authors wrote. “However, it has proven to be very difficult for hospitals independently to negotiate and enact single-vendor contracting for orthopaedic implant choice and use.” They noted that some orthopaedic surgeons seek to protect their prerogative to choose implant products and vendors and may resist implant choice dictated by hospitals or other external influences. Another challenge for the hospitals, the authors wrote, is that implant manufacturers have an advantage over hospitals in competing for the orthopaedic surgeons’ attention: “Surgeons are more likely to align with the vendor’s sales representative than with the hospital’s purchasing manager. This preferential alignment of the orthopaedic surgeon with the implant industry instead of the hospital is a major challenge to achieving value-driven health care.”
Between May 1, 2013, and Jan. 31, 2015, five surgeons agreed to follow up on patient outcomes to assess whether TKAs achieved equivalent quality when performed with either a POD-provided prosthesis or an industry-provided implant of their choosing. All knee implants used were approved by the FDA; Table 1 lists the standards of the American Association of Surgical Distributors. The POD sold two cemented TKA systems (Columbus, Aesculap; A200, Renovis Surgical) at a cost that was anticipated to be lower than the price used by the large industry companies (Zimmer, Biomet, Stryker, and Smith and Nephew). Surgeons selected either the POD products or the industry products at their own discretion without external influences, apart from each individual surgeon’s clinical decision-making and individual patient preference. All patients who underwent TKA with a POD implant were informed on the day of scheduling surgery that their surgeons had financial interest in using that implant, well before the actual date of surgery.
The Knee Injury and Osteoarthritis Outcome Score (KOOS) was used as the primary outcome measure. The researchers followed patients prospectively to determine KOOS outcomes at two years and any incidences of knee complications that required surgery.
No significant differences
A total of 596 knees met the final inclusion criteria. Of those, 209 (35.1 percent) had a POD knee implanted and 387 had an industry knee implanted at one of six local hospitals. Patients with POD knees were statistically significantly older (P = 0.05) than patients who received industry knees. However, the difference may not be of clinical significance because the average age difference was only two years (66 ± 8 versus 68 ± 10).
Both POD and implant knees showed statistically significant improvements (P < 0.0001) from baseline to final follow-up for all five subgroups of the KOOS, superseding the minimal clinically important difference (Fig. 1). No clinical or statistically significant differences were observed in any subgroup of the KOOS between the POD total knee prosthesis cohort and the industry total knee prosthesis cohort (Fig. 2). Knee complications requiring surgical intervention were similar between the groups (2.9 percent POD knees versus 3.6 percent industry knees; P = 0.58).
Because the study was not randomized and selection of knee prosthesis was at the surgeons’ discretion, there was the potential for selection bias, the authors noted. Therefore, it was important to identify whether surgeon, hospital, or implant factors could influence outcomes.
In examining the outcomes of POD knees versus industry knees stratified for each individual physician, the investigators observed no obvious trends identifying superiority of either group in mean improvements from baseline to final follow-up. In addition, when comparing outcomes of all total knee replacements among the five surgeons, the researchers observed no differences in outcomes for any of the KOOS subgroups: change in symptoms (P = 0.81), pain (P = 0.68), activities of daily living (P = 0.93), sport and recreation (P = 0.63), and quality of life (P = 0.13). “It also does not appear that an obvious predilection or bias was found for any of the surgeons to choose either the POD implant or the industry implant based on patients’ preoperative pain and function because no clear statistical trends were identified revealing differences in baseline KOOS subgroup scores between the groups,” the authors wrote.
The surgeries were performed at six different hospitals, but only two hospitals (n = 541 knees, 90.8 percent) were willing to share cost data. Surgeries on 444 knees were performed at Hospital 1. Of those, 209 were POD knees (47.1 percent), with an average cost of $4,067 per knee; 235 were industry knees (52.9 percent), with an average cost of $5,071.22 (P < 0.0001). Surgeries on 97 industry knees were performed at Hospital 2, with an average cost of $4,809. In that cohort, utilization of the lower-cost POD knees saved Hospital 1 a total of $209,876 compared to higher-cost industry knees.
Containing costs, not quality
“We were not surprised by the findings of our study,” Dr. Baumgarten said. “Currently, there is no evidence that any one company’s total knee prosthesis is clinically superior to any other prosthesis. Thus, it was ethical and appropriate to utilize the FDA-approved prostheses described in this study. Nevertheless, the members of the POD believed it was essential to track our clinical outcomes to ensure that the two prostheses groups provided similar benefits to our patients. In addition to the two-year outcomes reported in our manuscript, we examined KOOS scores and complication rates at the three-month time point to ensure that, if there was a difference between the prostheses groups, we would be able to identify this early and transition to 100 percent utilization of the superior knee prosthesis. As the study results show, the safety and efficacy of both prostheses were equivalent at two-year follow-up. We found similar results at the three-month follow-up, but we did not report these results in this manuscript.”
Limitations of the study, Dr. Baumgarten said, include the short duration of follow-up. “Obviously, this study did not compare longevity of the prostheses used, which is an important outcome to consider,” he said. Longer-term follow-up is planned for both the 10- and 20-year time points.
Overall, he said, “Physicians’ primary goal is to do their best to provide the safest and most effective treatments for their patients—the patient comes first. This is an exhaustive goal that requires time, attention, study, sacrifice, and practice. Cost containment is relegated secondary to patient care. Although physicians should be attentive to cost-containment strategies and practices for the greater good of society, it is rare that these strategies and practices directly affect (1) optimizing patient care or (2) physician reimbursement, which are typically the primary and secondary motivators for physicians. However, incentives change behaviors. It is likely that by further incentivizing physicians to educate themselves and practice cost-containment strategies, we may find more physicians finding the time and motivation to contribute to healthcare cost reduction.”
Dr. Baumgarten’s coauthors of “Do More Expensive Total Knee Arthroplasty Prostheses Provide Greater Improvement in Outcomes over Less Expensive Prostheses Sold by a Physician-owned Distributorship?” are Peter S. Chang, MD; Peter A. Looby, MD; Matthew J. McKenzie, MD; and Corey P. Rothrock, MD.
Terry Stanton is the senior science writer for AAOS Now. He can be reached at firstname.lastname@example.org.