The Pros and Cons of Modularity in Total Knee Replacements
Letters
and Replies
to the Editor of Journal of Bone and Joint Surgery American
To the Editor from Michael E. Berend, MD, Merrill A. Ritter, MD, The Center for Hip and Knee Surgery, Mooresville, IN.
A number of recent articles, including "In Vivo Deterioration of Tibial Baseplate Locking Mechanisms in Contemporary Modular Total Knee Components" (2001;83-A:1660-5), by Engh et al., and "The Effectiveness of Isolated Tibial Insert Exchange in Revision Total Knee Arthroplasty" (2002;84-A:64-8), by Babis et al., have discussed modularity of tibial components in total knee arthroplasty. Both of these studies mentioned the potential advantage of a reduction of implant inventory with use of modular components.
We have not experienced these perceived benefits of modular tibial components in total knee arthroplasty. We currently have an inventory of both modular and nonmodular total knee arthroplasty systems. The nonmodular system that we use is the AGC (Anatomic Graduated Components) Total Knee System (Biomet, Warsaw, Indiana). To have an adequate inventory for use in primary total knee arthroplasty, we maintain three patellar sizes; five tibial component sizes, each with seven corresponding polyethylene thicknesses; and five universal femoral components, which can be used in right or left total knee replacements. Because the system is nonmodular, a total of forty-three boxes of implants must be stocked and available for a primary right or left knee replacement. We also have an inventory of modular knee replacement components from the Zimmer NexGen Legacy LPS system (Zimmer, Warsaw, Indiana). This system includes five patellar sizes, six tibial trays with six available polyethylene thicknesses per tray, and five anatomic femoral components (a total of ten femoral components). This amounts to a total of fifty-one boxes of implants that must be available in inventory for a primary right or left total knee arthroplasty.
Potential disadvantages of modularity of the tibial component include backside polyethylene wear and limited effectiveness and early failure of isolated tibial insert exchange. On the basis of both our clinical results and an actual reduction in implant inventory with use of nonmodular components, we continue to utilize primarily nonmodular tibial components in primary total knee arthroplasty.
Reply
from Gerard A. Engh, Anand R. Rao, and Matthew B. Collier, Anderson
Orthopaedic Research Institute, Alexandria, VA:
We appreciate the opportunity to respond to the letter by Drs. Berend
and Ritter concerning our article. They raised questions about the potential
advantages of modularity in regard to inventory reduction. Although
we agree with their conclusion, we found their argument flawed. They
constructed a scenario comparing a nonmodular system with a modular
system: the nonmodular system used five sizes of universal femoral components
(no right or left components), three sizes of patellar components, and
five sizes of tibial components with seven different thicknesses; the
modular system had right and left femoral components, five sizes of
patellar implants, and six sizes of tibial components.
It is true that the nonmodular system in their example necessitated less inventory than did the modular system, but the writers did not address the advantages of the modular system in providing right and left femoral components, various sizes of patellar implants, and the ability to use a larger-size tibial tray with different femoral components.
The real advantages of a modular system are interchangeability of different sizes of femoral and tibial components, the opportunity to adjust component thickness prior to closure, the ease of implant placement, and access to the back of the knee for hemostasis and removal of extruded cement. Furthermore, although the number of components in inventory were equivalent for the two systems in their example, Drs. Berend and Ritter did not address the fact that a nonmodular system requires more storage space, since each component is the size of a full tibia.
Despite my disagreement with the example offered by Drs. Berend and Ritter, I do agree with their choice of nonmodular tibial components for use in most primary total knee arthroplasties. From the standpoint of component wear, modularity represents a compromise. Eliminating tibial insert motion remains a difficult engineering challenge because of the marked difference in the modulus of elasticity between polyethylene and metal. Backside wear is a significant source of wear debris that can be eliminated simply by selecting nonmodular components for use in primary total knee arthroplasty.
Reply
from B.F. Morrey, G.C. Babis,
and R.T. Trousdale:
In regard to the letter by Drs. Berend and Ritter, we completely agree
with their position. In regard to the discussion of the rationale for
the development of modular implants, we also mentioned in our article
the potential for decreased inventory due to the interchangeability
of parts. However, we completely agree that, in fact, modularity has
increased rather than decreased the inventory carried by hospitals when
compared with that necessary for monoblock implants. The thrust of the
review, of course, was to demonstrate that the presumed value of interchangeable
parts in enabling revisions to be more readily accomplished has not
been documented, in our experience.