Manual instruments in total knee arthroplasty (TKA) size the femur from a fixed reference point, either the anterior femoral cortex or the posterior femoral condyles. Posterior-referenced systems are said to ensure consistent restoration of posterior femoral offset1 and avoid inappropriate tensioning of the resulting flexion gap.2 Whereas “kinematic” TKA alignment aims to reproduce the position of both posterior femoral condyles with the implant,3 posterior-referencing instruments designed to achieve “classic” TKA alignment with a measured-resection technique allow the surgeon to externally rotate the femoral prosthesis relative to the posterior condylar axis.4 The goal is to align the prosthesis with the flexion–extension axis of the knee as approximated by the transepicondylar axis,5–8 but this choice makes it impossible to anatomically restore the posterior femoral offset of both condyles simultaneously.
A measured-resection technique is typically assisted by instrumentation that allows the surgeon to re-create the transepicondylar axis by directly measuring the posterior femoral condylar axis with reference paddles and adding a certain amount of external rotation to that axis, as the posterior condylar axis typically lies in slight internal rotation to the transepicondylar axis.
When the femoral prosthesis is externally rotated relative to the posterior condylar axis, a geometric axis must exist around which this rotation is accomplished (Figure 1A–C). The location of this axis of rotation—medial, lateral, or in between—will have implications for the resulting posterior resections and thus the resulting restoration of posterior offset of each condyle. The instrumentation for any given posterior-referencing system will determine the location of the axis of rotation. A medial axis would be expected to restore the offset of the medial femoral condyle and increase the offset of the lateral femoral condyle with increasing external rotation, lowering the posterior joint line. A lateral axis would be expected to restore the offset of the lateral condyle and decrease the offset of the medial condyle with increasing external rotation, raising the posterior joint line. A central axis would be expected to result in balanced changes in the offset of the medial and lateral condyles with negligible effect on the level of the joint line as measured at the midpoint between the two femoral condyles.
Representation of lateral (A, D), center (B, E), and medial (C, F) posterior condylar reference points for varying femoral component external rotations. The 0° line represents a bone resection parallel to the posterior condylar axis. Abbreviations: A, anterior; L, lateral; M, medial; P, posterior.
Changes in the joint line can have a significant influence on knee stability postoperatively.9 Absent preoperative imbalance between flexion and extension laxity, changes in the distal femoral joint line should match changes in the posterior femoral joint line caused by femoral rotation to achieve balanced flexion and extension gaps. Most distal femoral cutting jigs are designed to reference the most distal femoral condyle, whether it be medial or lateral. Due to the frequent presence of distal femoral valgus relative to the mechanical axis, even in varus knees, this is most often the medial condyle. This would be expected to lower the joint line laterally in cases where the native valgus of the distal femur exceeds the planned valgus of the femoral implant, which is often the case in classic TKA alignment.
If the medial joint line is referenced in extension, jigs that fail to reference the medial joint line in flexion may result in differential changes to the flexion and extension joint lines. This may, in turn, contribute to flexion laxity or extension tightness, depending on the magnitude of tibial resection.10 Such discrepancy may be particularly significant in posterior-stabilized TKA, as resection of the posterior cruciate ligament may cause further flexion laxity and compound any imbalance.11
To the current authors' knowledge, this phenomenon has not been comprehensively described in the literature, but it should be understood by any surgeon hoping to efficiently use posterior-referencing instrumentation to achieve balanced flexion and extension gaps in TKA. To avoid mismatch between the flexion and extension gaps, surgeons may wish to be cognizant of which joint line is restored in extension and endeavor to restore the posterior offset of the corresponding condyle in flexion (or vice versa).
The authors set out to quantify, in a controlled laboratory setting, how femoral external rotation affects restoration of posterior condylar offset with various contemporary TKA systems, depending on jig design and specifically rotation axis. They then developed practical guidelines to mitigate the unintended consequences of femoral rotation on the balance of flexion and extension gaps.
Materials and Methods
The authors examined 9 different posterior-referenced TKA instrumentation systems available at their institution: Legion and Journey II (Smith & Nephew, Andover, Massachusetts), Persona and NexGen (Zimmer, Warsaw, Indiana), Sigma and Attune (DePuy Synthes, West Chester, Pennsylvania), Vanguard (Biomet, Warsaw, Indiana), Triathlon (Stryker, Kalamazoo, Michigan), and Unity (Corin, Tampa, Florida) (Table 1). The instrumentation systems were examined to determine whether femoral rotation occurred around a medial, a lateral, or an intermediate reference point (Figure 1D–F).
Implant Systems Examined
Of the 9 systems examined, 8 have symmetric prosthetic medial and lateral femoral condyle thicknesses. One of these systems (Legion) uses a laterally referenced jig to set femoral rotation and 1 (Unity) uses a medial reference, whereas 6 systems use a central reference point. One implant system (Journey II) has asymmetric medial and lateral femoral condyles designed to create an oblique (and theoretically more anatomic) joint line despite bony resection according to mechanical alignment principles. This system was included in this study because it is one of only two systems identified to use a medial reference for femoral rotation.
Using identical left distal femoral Sawbones (Pacific Research Laboratories, Vashon Island, Washington), with manufacturer-stated specifications of 5.5° to 6° distal femoral valgus angle and 3° of external rotation of the transepicondylar axis in relation to the posterior condylar axis, manual instrumentation from each system was used to perform standard distal and posterior femoral resections. Four Sawbones trials were performed per implant system. All distal femoral resections were made at 5° of valgus alignment to the intramedullary axis using the system's alignment jig (Figure 2). Two sets of posterior femoral resections were performed with the posterior-referencing jigs from each system set at 3°, and then two more were performed at 5° of external rotation. The same oscillating saw and 1.27-mm saw blades were used for all resections. The distal and posterior femoral resections for each trial were measured at their widest dimension with a standard caliper, adding the width of the saw blade to estimate and account for unmeasured bone removal. These values were recorded and averaged for each setting. All study procedures were performed by or directly supervised by an attending surgeon (M.S.H.) with fellowship training in adult reconstructive hip and knee surgery.
Distal femoral cutting jig applied to distal femur at 5° valgus setting.
Using manufacturer-provided specifications (Table 2) to determine the thickness of the prosthetic femoral condyles distally and posteriorly, the authors then calculated changes in the distal medial, distal lateral, posteromedial, and posterolateral femoral joint lines for each TKA system at both 3° and 5° of external rotation. The authors' outcomes of interest were (a) magnitude of any changes in the level of the distal or posterior femoral joint lines as measured medially and laterally and (b) magnitude of mismatch between changes in the distal and posterior joint lines.
Manufacturer-Provided Condylar Thicknesses of the Femoral Implants
Direct measurements of the femoral sizing jigs at 3° and 5° settings of external rotation also were taken using a metal ruler. Distances were measured from the posterior paddles to the medial and lateral reference drill holes used to set the position of the subsequently placed 4-in-1 femoral cutting jig. The difference between the medial distance and the lateral distance was designated the predicted difference between posteromedial and posterolateral resection magnitudes for a given jig at a given setting. The purpose of this measurement was to determine whether predicted differences in resection magnitudes would corroborate the authors' observed differences in resection magnitudes on the Sawbones models. If the difference between 2 Sawbones trials at the same setting was far smaller than the differences between trials at different settings, and if the observed difference in resection magnitude could be predicted by direct measurement of the jigs themselves, additional Sawbones trials would not be necessary to confirm that the primary source of variation between trials was jig setting and not jig imprecision or surgical technique.
For the authors' laterally referenced system, Legion, the posterolateral femoral joint line was restored to within 0.2-mm accuracy at both 3° and 5° of external rotation (Table 3). However, a −4.2-mm medial-sided discrepancy (flexion over-resection) was seen between the flexion and extension joint lines when 5° of external rotation was selected, as might be required in a valgus knee with a hypoplastic lateral femoral condyle. The 3° of external rotation commonly selected for routine varus knees resulted in a smaller, −2-mm medial mismatch between the distal and posterior joint lines. Lateral joint line discrepancies between flexion and extension were smaller in magnitude: 0.9 and 1.1 mm at 3° and 5° of external rotation, respectively.
Change in Offset Restoration as External Rotation Was Increased From 3° to 5°
The authors' 6 center-referenced systems exhibited a −2.4-mm discrepancy between the flexion and extension joint lines medially with 5° of external rotation (Table 3), while a smaller, −1.5-mm discrepancy was noted with 3° of external rotation. Measured lateral discrepancies were slightly smaller in magnitude: +0.6 mm at 3° of rotation and +1.4 mm at 5° of rotation.
Of the 2 medially referenced systems, 1 system, Unity, has symmetric posterior condylar thicknesses, while the other, Journey II, has asymmetric posterior condylar thicknesses by design (Table 2). Both systems exhibited absolute medial flexion–extension offset discrepancies of less than 0.6 mm for all settings of external rotation. At 5° of external rotation, the symmetric system exhibited a lateral discrepancy of +4.3 mm, and the asymmetric system exhibited a discrepancy of +2.9 mm. At 3° of external rotation, lateral discrepancies were smaller.
Direct measurements of the femoral sizing jigs at 3° and 5° settings of external rotation revealed that the predicted difference between posteromedial and posterolateral resections varied in a predictable fashion based on the jig design. With increasing external rotation, the laterally and medially referenced jigs demonstrated little to no change in the referenced side's distance from posterior paddle to ipsilateral drill hole, indicating no change in predicted resection magnitude. For these jigs, predicted resection magnitude changed 2 to 3 mm on the non-referenced side when external rotation was increased from 3° to 5°. Measurement of the center-referenced jigs predicted, on average, approximately 1 mm more resection medially and 1 mm less resection laterally as external rotation was varied from 3° to 5°, corresponding to over-resected bone (under-restored offset) medially and under-resected bone (over-restored offset) laterally. All predictions corresponded to what the authors observed in their resection trials (Table 3).
The authors examined the jigs from 9 modern total knee systems to understand the effect of jig design on restoration of distal and posterior femoral joint lines in TKA, paying particular attention to the effect of the axis about which femoral external rotation occurs.
Distal femoral cutting jigs from all systems tended to contact the distal femoral condyles medially rather than laterally. These systems tended to increase distal lateral femoral offset while maintaining the distal medial joint line within 0.3 mm (Figure 3). This is unsurprising, as the Saw-bones examined had a distal femoral valgus angle of 5.5° to 6°, meaning that a 5° valgus jig would be expected to reference medially and not laterally. Minor observed variability is likely a result of (1) the imprecision of any intramedullary guide when the intramedullary rod is smaller than canal diameter and (2) the inherent imprecision of the saw–jig–bone interface.
Distal femoral joint line alteration by reference point location. Over-restoration refers to a joint line that has been lowered, whereas under-restoration refers to a joint line that has been raised. Default medial (DM) femoral referencing on all systems produces an equal bony resection on all 3 groups. Abbreviation: DL, distal lateral.
The center-referenced systems showed trends toward both posteromedial offset under-restoration and posterolateral offset over-restoration. However, the overall magnitudes of joint line alterations were not as great as those seen in the medial- and lateral-referenced systems (Figure 4).
Posterior femoral offset alteration (increased or decreased offset) at 3° of external rotation setting. Referencing laterally produces a significant bony resection off of the medial condyle. Referencing medially produces only a minimal change in offsets using an asymmetric implant. Abbreviations: PL, posterior lateral; PM, posterior medial.
In the laterally referenced system, posterolateral femoral offset was accurately restored, while posteromedial offset was reduced. The magnitude of under-restoration increased with greater external rotation (Figures 4–5). The differential between a medial reference in extension and a lateral reference in flexion resulted in lowering the extension joint line and raising the flexion joint line, with the potential for clinically important imbalance of the flexion and extension gaps with increasing external rotation.
Posterior femoral offset alteration (increased or decreased offset) at 5° of external rotation setting. Referencing laterally produces a significant increase in bony resection off of the medial condyle. Referencing medially at 5° of external rotation produces a significant increase in lateral condyle offset. Abbreviations: PL, posterior lateral; PM, posterior medial.
Both medially referenced systems, 1 (Unity) with symmetric and 1 (Journey II) with asymmetric condylar thicknesses—designed to ideally create a more anatomic oblique joint line—achieved accurate posteromedial femoral offset restoration at both rotational settings but yielded posterolateral over-restoration at 5° of external rotation (Figure 5) by approximately one femoral size.
Although the general trends observed were predictable based on a thorough inspection of the jigs examined, the magnitudes of differences measured both demonstrate the importance of understanding the jig being used and provide guidance for strategies to achieve gap balance with each jig type.
The authors' results confirm their clinical impression that the laterally referenced system's instrumentation tends to reduce posterior femoral condylar offset. This resulted in flexion–extension joint line mismatch of up to 4.2 mm at 5° of external rotation. To prevent under-restoration of posterior femoral offset with such jigs, the authors recommend a routine posterior shift of 1 to 2 mm when 2° to 3° of external rotation is selected and a routine posterior shift of 2 mm when 4° or more of rotation is selected. If external rotation is 5° to 6°, an additional 1- to 2-mm distal femoral resection may be considered to balance the flexion and extension gaps. Although an additional 2-mm posterior shift could obviate the need for additional distal femoral resection, a 4-mm posterior shift would result in selection of a larger femoral size, which might overhang medially and/or laterally. As posterior cruciate ligament tension and tibial slope play a role in flexion gap tightness,12 a posterior shift at the lower end of the recommended range may be advised if the posterior cruciate ligament is being preserved and/or the tibial slope is being reduced, whereas a posterior shift at the larger end of the range may be advised if the posterior cruciate ligament is being substituted and/or the tibial slope is being maintained or increased.
Surgeons using center-referenced systems (Attune, Sigma, NexGen, Persona, Vanguard, and Triathlon, among others not studied) should be aware that the posteromedial offset is reduced by greater than 2 mm when 5° of external rotation is selected. This may justify a posterior shift of 1 to 2 mm when the jig allows, a reduction in the slope of the tibial cut, or performance of an additional 1- to 2-mm distal femoral resection to balance the gaps. The additional 1- to 2-mm distal femoral resection will also prevent excessive lowering of the distal lateral femoral joint line in severely valgus knees, and may be the optimal strategy in that setting.
The medially referenced system with asymmetric condylar thicknesses is relatively unique. A surgeon aware that this medially referenced system can over-restore posterolateral femoral offset by 3 mm with a 5° external rotation setting might elect to anteriorize the femoral component in an effort to avoid a tight flexion gap, but should recall that this system references the medial condyle to set the extension joint line in a valgus knee. The standard instrumentation will reference the lateral joint line in extension only when there is varus deformity of the distal femur, which in the authors' experience rarely correlates with the need for 5° of femoral external rotation. This maneuver may occasionally be useful when the posterior cruciate ligament is preserved in a varus knee with unusual rotational anatomy where reduction in posteromedial offset could help ligament balance. Conversely, decrease in the posteromedial offset in a valgus knee with a posterior cruciate ligament–sacrificing design could result in flexion laxity. When using the cruciate-substituting version of the implant studied, the authors anteriorly shift the jig only if required to match additional distal femoral resection performed to achieve adequate lateral bone–prosthesis contact in the setting of a knee with extreme distal femoral valgus.
One important limitation of this study was that the controlled Sawbones setting limited the external validity of the findings and did not reflect the diversity of clinical scenarios potentially encountered in practice with varying bone morphology. Nevertheless, the results provide quantification of the differences between jigs not distorted by variables related to adequacy of exposure, bone quality, and osteophyte impingement that could confound in vivo measurement of these data. This made for high internal validity and allowed the authors to confidently describe how femoral rotation affects the posterior joint line with each jig design. Furthermore, having recognized the expected adverse consequences of using each jig as designed rather than altering the level of resection to balance the flexion and extension gaps, the authors believed it would have been unethical to have performed this study in vivo.
A potential criticism of this study is that only 4 Sawbones were used per system, with only 2 trials at each rotation setting. Although standardized synthetic bones of identical morphology were used, one might expect a certain amount of variability between trials because of inherent imprecision in the mechanical jigs and power tools used. For each system, the expected difference in magnitude between posteromedial and posterolateral resections at a given degree of external rotation was determined by directly measuring the distance between the medial posterior-referencing paddle and the medial pin hole that sets cutting jig placement and comparing that with the distance between the lateral posterior-referencing paddle and the lateral pin hole that sets cutting jig rotation. Then, the authors compared their observed differences in resection magnitude with those predicted by direct jig measurement. For both posteromedial (r=0.79) and posterolateral (r=0.84) resections, correlation in the observed resection magnitude change from 3° to 5° was strong, indicating that the authors' jigs produced observed changes in resection in the pattern predicted based on direct jig measurement. Correspondingly, the authors did not believe that additional Sawbones trials would have been useful in illustrating the fact that jig design can cause flexion–extension mismatch at higher degrees of femoral component external rotation.
Finally, as the authors' goal was only to describe and quantify the performance of various mechanical femoral cutting jigs, this study did not aim to quantify the effects of various soft tissue–related balancing methods such as ligament releases on imbalance between flexion and extension gaps, nor did it examine the effect of tibial cut slope or alignment on final balance. Soft tissue tensioning plays an important role in balancing of flexion and extension gaps in both gap-balancing and measured-resection surgical techniques. This study is more relevant for the surgeon using a measured-resection technique, as bony resections are the foundation of knee stability and balance in this method, and ligament releases are conducted secondarily to achieve final balance. An unintended excess or inadequate posterior femoral resection will lead to a bony imbalance in flexion and extension gaps that will likely not be correctable by intentional soft tissue releases. Release or retention of the posterior cruciate ligament will affect flexion gap tightness, but this decision may be made earlier in the operation. Indeed, it is not possible to restore a posterior cruciate ligament that has already been fully released if one desires to tighten the flexion gap. Additionally, elevation of the joint line has been associated with midflexion instability in the literature,13 so excess femoral resection that is compensated for by building up the tibial component could create further unintended instability.
Variability in TKA posterior-referencing jig design results in variation in posterior condylar offset restoration. This can result in differential changes in the distal and posterior femoral joint lines after TKA. Surgeons should familiarize themselves with the design of the jigs used for their implant of choice and determine whether there are scenarios in which routine changes in jig position are advisable to achieve balanced gaps. Furthermore, any surgeon changing implant systems should become particularly attuned to differences in jig design between systems. Applying strategies that are successful with one jig design to an implant with a different jig design could result in unexpected difficulties balancing the soft tissue envelope. Future TKA implants and posterior-referencing jigs should be designed with attention to restoring the same joint line with distal and posterior femoral resections.
- Bellemans J, Banks S, Victor J, Vandenneucker H, Moemans A. Fluoroscopic analysis of the kinematics of deep flexion in total knee arthroplasty: influence of posterior condylar offset. J Bone Joint Surg Br. 2002;84(1):50–53. doi:10.1302/0301-620X.84B1.0840050 [CrossRef]11837832
- Bong MR, Di Cesare PE. Stiffness after total knee arthroplasty. J Am Acad Orthop Surg. 2004;12(3):164–171. doi:10.5435/00124635-200405000-00004 [CrossRef]15161169
- Howell SM, Howell SJ, Kuznik KT, Cohen J, Hull ML. Does a kinematically aligned total knee arthroplasty restore function without failure regardless of alignment category?Clin Orthop Relat Res. 2013;471(3):1000–1007. doi:10.1007/s11999-012-2613-z [CrossRef]
- Griffin FM, Insall JN, Scuderi GR. The posterior condylar angle in osteoarthritic knees. J Arthroplasty. 1998;13(7):812–815. doi:10.1016/S0883-5403(98)90036-5 [CrossRef]9802670
- Whiteside LA, Arima J. The anteroposterior axis for femoral rotational alignment in valgus total knee arthroplasty. Clin Orthop Relat Res. 1995;(321):168–172.7497664
- Churchill DL, Incavo SJ, Johnson CC, Beynnon BD. The transepicondylar axis approximates the optimal flexion axis of the knee. Clin Orthop Relat Res. 1998;(356):111–118. doi:10.1097/00003086-199811000-00016 [CrossRef]
- Berger RA, Rubash HE, Seel MJ, Thompson WH, Crossett LS. Determining the rotational alignment of the femoral component in total knee arthroplasty using the epicondylar axis. Clin Orthop Relat Res. 1993;(286):40–47.8425366
- Hepinstall MS, Ranawat AS. Landmarks for optimizing component position in total knee arthroplasty. Curr Orthop Pract. 2008;19(2):147–154. doi:10.1097/BCO.0b013e3282f5400c [CrossRef]
- Martin JW, Whiteside LA. The influence of joint line position on knee stability after condylar knee arthroplasty. Clin Orthop Relat Res. 1990;(259):146–156.2208849
- Parratte S, Pagnano MW. Instability after total knee arthroplasty. J Bone Joint Surg Am. 2008;90(1):184–194.18172954
- Dennis DA, Komistek RD, Scuderi GR, Zingde S. Factors affecting flexion after total knee arthroplasty. Clin Orthop Relat Res. 2007;464:53–60.17621230
- Jojima H, Whiteside LA, Ogata K. Effect of tibial slope or posterior cruciate ligament release on knee kinematics. Clin Orthop Relat Res. 2004;(426):194–198. doi:10.1097/01.blo.0000138960.57680.60 [CrossRef]15346073
- Ramappa M. Midflexion instability in primary total knee replacement: a review. SICOT J. 2015;1:24. doi:10.1051/sicotj/2015020 [CrossRef]27163080
Implant Systems Examined
|Systema||Refb||Jig-M,c mm||Jig-L,d mm|
Manufacturer-Provided Condylar Thicknesses of the Femoral Implantsa
Change in Offset Restoration as External Rotation Was Increased From 3° to 5°
|System||DM-O,a mm||DL-O,a mm||PM-P,b mm||PM-O,b mm||PL-P,c mm||PL-O,c mm|