The overarching goals of total knee arthroplasty (TKA) are pain reduction and limb realignment if a deformity is present, while both patients and surgeons strive for joint stability.1–4 In addition, implant survivorship of TKA has been associated with balancing the soft tissues around the joint.5,6

Unfortunately, there is a lack of consensus in the orthopedic community regarding the appropriate postoperative limb alignment after TKA.7,8 This variability may contribute to the disparity between the patient-reported satisfaction rates for THA in comparison to TKA, as hip replacement satisfaction continues to consistently score higher than TKA satisfaction.9–12

The purpose of this study is to report on the outcomes associated with a newly devised surgical technique that was developed with the following goals: simplicity, reproducibility, and improved patient satisfaction. This novel surgical technique is presented with the intent of balancing coronal malalignment in modern TKA without soft tissue releases, computer-assisted surgery, or robotics.

MATERIALS AND METHODS Overview

The clinical outcomes of 433 TKAs performed in 337 patients were collected retrospectively in an Institutional Review Board (IRB) approved study. All consecutive patients were included in the study cohort. The surgeries were performed by a single surgeon at 3 facilities: a level 1 trauma center, an orthopedic subspecialty hospital, or same-day home surgery at an ambulatory surgery center. Patients received a combination of cruciate retaining (CR) and posterior stabilized (PS) implants from various implant companies. Each patient’s posterior cruciate ligament was assessed intraoperatively, as a tibia-first technique was used. If the posterior cruciate ligament showed continuity, then a CR implant was used; however, if the posterior cruciate was clinically absent or insufficient after tibial resection, then a PS design was implanted. No highly constrained or hinged implant designs were utilized. All surgeries were performed under a tourniquet, all implants were cemented, all patellae were resurfaced, and no drains were utilized. All patients received perioperative antibiotics. Patients were seen for a preoperative visit, a 2- and 6-week follow-up, then an annual evaluation.

Radiographic Criteria

Preoperative and postoperative long-leg imaging was available for the study group once the principal investigator (PI) transitioned to digital radiography; therefore, the surgeries chosen for this study were performed between November 2016 and August 2019. The hip, knee, ankle (HKA) angle of the radiographic coronal alignment was measured according to the method as described by the Knee Society13 (Figs. 1A, B). A board-certified radiologist performed the measurements using Medstrat Software (Medstrat Inc. 1901 Butterfield Rd. #600 Downers Grove, IL, ECHOS version 3.2.18.0).

FIGURE 1:

A, Severe valgus (20.4 degrees left knee) and varus (10.6 degrees right knee) malaligned knees, a so-called “wind-swept” deformity preoperatively. B, Postoperative radiographs of (A) showing improved coronal alignment (20.4 to 7.3 degrees valgus left knee and 10.6 to 2.3 degrees varus right knee) and patient-reported outcomes revealed “back to normal” (level of improvement: 4) and “completely satisfied” (patient satisfaction: 6).

Surgical Technique

The surgical exposure was created through a medial parapatellar approach without any additional soft tissue releases. The first step in the technique was to remove the marginal osteophytes that surround the distal femur and proximal tibia (Fig. 2). Then, an extramedullary tibia-first technique was used. An attempt was made to resect the tibia at 0 degree of varus/valgus with a 3-degree posterior slope as this was the angle of the manufacturer’s cutting guide.

FIGURE 2:

Displaced medial collateral ligament secondary to “tenting,” not contracture, can return to its native position after the removal of marginal osteophytes and the meniscus. Arrow pointing at tenting medial collateral ligament.

Next, the exposed proximal tibial metaphyseal bone was protected with a metallic plate (Fig. 3). Then, a novel linked and tensioned extramedullary resection (LATER) technique was utilized. More specifically, the collateral ligaments are tensioned equally, medially and laterally, with calibrated Scott femoral tibial tensioners (Innomed Inc., Savannah, GA) to a measured amount of ~25 to 30 ft/lb (2.5 to 3 indicator gauge on the tensioners) in 0 degree extension (Fig. 4A). Once the extension gap was tensioned, a newly developed jig called the Missing Link Device (Sterling Innovations LLC, Scottsdale, AZ) was used to link the previously created tibial resection to affix the distal femoral resection block at the level of the femoral notch with the limb in full extension (Fig. 4B). The Missing Link Device was disconnected, the knee was flexed, and the distal femoral resection was completed without intramedullary violation. This created a balanced rectangular extension gap without any soft tissue releases.

FIGURE 3:

Flat metallic protective plate for the resected proximal tibial exposed metaphyseal bone.

FIGURE 4:

A, Equally tensioned medial and lateral compartments while the knee is fully extended. B, With the linked and tensioned extramedullary resection technique the Missing Link Device connects the anticipated distal femoral resection to the previously created proximal tibial resection resulting in a balanced rectangular extension gap without the traumatic release of soft tissues.

From the sagittal plane, the newly created rectangular extension gap is actually a trapezoidal space, not rectangular; thus, standard spacer blocks would have the surgeon assessing mediolateral stability in a 3-degree flexed position, instead of full extension. Therefore, a unique trapezoidal spacer block was created to allow assessment of the extension gap in full extension at 0 degree (Fig. 5).

FIGURE 5:

Sagittal schematic diagram showing the femur and tibial create a trapezoidal shape when compared with the rectangular shape created in the coronal plane. A specialized spacer block with a 0 degree shim on the femoral side and a 3 degree shim on the tibial side was used to confirm M/L stability in full extension and flexion.

The extension gap was measured in mm per the appropriate spacer block (typically 17 to 23 mm), then the same tension was applied medially and laterally in the flexion space at 90 degrees to produce external rotation (ER) of the femoral component using posterior referencing gap balancing technique to create the rectangular flexion gap that matched the size and shape of the extension gap. Several implant companies offer instrumentation that allow tensioning while the knee is flexed (Fig. 6).

FIGURE 6:

Examples of flexion gap balancer instrumentation as offered from DJO and Exactech, respectively.

Data Collection

We then evaluated postoperative long-leg radiographs to reassess HKA angular alignment during the 2-week follow-up visit.

We collected patient demographic data such as age, gender, and body mass index. We also looked at several other inpatient and outpatient parameters such as: opioid utilization by reviewing the state-run controlled substance prescription monitoring program, length-of-stay, 30-day readmissions, and any adverse outcomes. Finally, KOOS Jr, patient satisfaction rates and levels of improvement were assessed using 6-point and 4-point Likert scales, respectively (Tables 1, 2).

TABLE 1 - Four-point Likert Scale Used for the Determination of Patients’ Level of Improvement After Total Knee Arthroplasty Level of Improvement “How Much Has the Knee Replacement Improved Your life?” 0 No improvement or worse 1 A little improvement 2 A moderate amount of improvement 3 A lot of improvement 4 Back to normal/complete improvement
TABLE 2 - Six-point Likert Scale Used for the Determination of Patients’ Satisfaction After Total Knee Arthroplasty Patient Satisfaction “How Satisfied Are You With Your Knee Replacement?” 0 Completely dissatisfied 1 Mostly dissatisfied 2 Somewhat dissatisfied 3 Neither dissatisfied nor satisfied 4 Somewhat satisfied 5 Mostly satisfied 6 Completely satisfied
RESULTS

Of the 433 TKAs, 89 had incomplete data (59 incomplete radiographic visualization of the lower extremity, 26 lost to follow-up, 3 deceased, and 1 could not stand for preoperative imaging), leaving 344 TKAs with complete data. Two hundred ten had normal alignment (<10 degrees varus/valgus) and 134 were malaligned. Forty-eight patients underwent bilateral surgeries within the study period, of which 31 had complete data (62 TKAs). Three hundred twelve procedures were performed in an inpatient hospital setting, 31 in an outpatient ambulatory surgery center setting, and 1 in an orthopedic specialty hospital. The clinical follow-up ranged from 12 to 46 months. The mean age at the time of surgery was 67 and ranged from 46 to 94 years old. The average body mass index in the normal alignment group was 31.7 and was similar in the malaligned group at 31.2. (range: 21.7 to 44.0). One hundred forty-two knees received a CR designed implant, the remaining 202 knees received a PS implant. Patients received implants from ZimmerBiomet (Warsaw, IN), Exactech (Gainesville, FL), Smith & Nephew (Memphis, TN), Johnson & Johnson/DePuy (Warsaw, IN), and DJO (Austin, TX). There were no transfusions, nor neurovascular injuries.

The final cohort had 258/344 preoperative varus deformity. The severe preoperative deformities (>10 degrees) were unequally distributed by sex, 59% were women; however, 65% of the entire 344 TKAs were performed in women. The average preoperative alignment in the normal cohort was 3.2 degrees (varus) which resulted in an average postoperative alignment of 1.0 degrees (varus). Whereas, the mean preoperative angular valgus malalignment was 13.8 degrees (range: 10 to 26.6 degrees deformity) and the varus malalignment was 12.3 degrees (range: 10 21.2 degrees deformity). The resultant mean postoperative alignment in the malalignment subgroups was 5 degrees valgus and 1 degrees varus, respectively.

The average coronal angular correction in the normal alignment cohort was 2.2 degrees; whereas, the mean coronal alignment correction in the malaligned group was 8 and 13 degrees, in the valgus and varus cohorts, respectively. Bilateral TKA patients’ preoperative alignment difference side-to-side averaged 4.4 degrees (range: 30.4 to 0 degrees), while postoperative alignment was found to be within 2.9 degrees of each other (range: 10 to 0.2 degrees). The mean angular correction with the LATER technique was 8.2 degrees (Figs. 7A, B).

FIGURE 7:

A, Negative numbers represent a varus malalignment, while positive integers are equated to a valgus malalignment. B, Negative numbers represent a varus alignment correction, while positive integers are equated to a valgus alignment correction. HKA indicates hip, knee, ankle.

We were able to obtain patient-reported outcomes in 249/344 (72.3% response rate). Overall patient satisfaction was found to be 89.2%. Moreover, 92.3% of patients reported “a lot of improvement” or “back to normal.” Ten of 344 patients reported dissatisfaction secondary to persistent pain, tightness, and stiffness. The average raw KOOS Jr. score obtained at a minimum of 1 year follow-up was 3.6 which equated to an adjusted score of 81.8.

There were no revision arthroplasties performed on this study group for loosening, wear, nor instability; however, we did encounter adverse outcomes in a small group of patients. Three patients were noted to have readmissions within 30 days postoperatively. More specifically, 1 patient had metabolic acidosis secondary to rhabdomyolysis and underlying liver cirrhosis. The other 2 recovered fully from a gastrointestinal bleed and leukopenia, respectively. Meanwhile, 5 patients did undergo additional knee surgery. First, the patient with liver cirrhosis cited above also developed persistent postoperative bleeding secondary to a coagulopathy, which led to wound healing issues. This same patient then developed an infection, was treated with a 2-stage revision, had recurrent infection and ultimately an above the knee amputation. Four of the remaining patients sustained disruption of the medial arthrotomy repair postoperatively, 3 within the first 90 days and 1 chronic addressed surgically over a year postoperatively for patellar realignment. Finally, 1 patient underwent manipulation under anesthesia postoperatively and ultimately achieved 0 to 125 degrees range of motion.

DISCUSSION

The purpose of this study is to present a novel surgical method that can improve both radiographic alignment (an average of 8.2 degrees correction) and patient-reported outcomes (>89.2% improvement and >79.3% satisfaction rates) while simplifying the execution of modern TKA. We have found that the data supports the following: (1) the technique is applicable to both CR and PS knee implant designs; (2) the technique is open-platform, meaning that it can be used universally with several contemporary knee replacement systems; (3) a balanced knee replacement construct is created without ligament releases in malaligned arthritic knees; (4) there is decreased blood loss when compared with IM-guided TKA alignment and (5) finally, there is improved patient-reported outcomes when specifically looking at overall satisfaction and improvement comparable to that of THA.

Multiple implant companies were used during this study for varying reasons: match the contralateral implant, a patient reported history of metal sensitivity, or facility directed implant availability. This strengthens the concept that this is indeed an open-platform technique.

Failure mechanisms for TKAs have shifted from implant-related issues, such as polyethylene wear, to instability and malalignment. Thiele and colleagues found that in 358 revision TKAs, the cause for additional surgery 42.5% of the time was either instability (21.8%) or malalignment (20.7%).14 Because of this, the classic mechanical alignment (MA) method of TKA has been challenged by the kinematic alignment (KA) as presented by Howell who reports a decreased length of stay 1.3 days versus 3.1 days for MA-TKA; lower 30-day readmission 1.5% compared with 2.5%; and, 1% versus 2.7% major complications.15

Although there are differences between the MA-TKA and KA-TKA, the commonality of the tibia and femur being addressed in isolation is similar for both techniques. This also applies to the adoption of any current robotic or navigated methods to perform contemporary TKAs. Whereas, the LATER technique differs from these methods because it uses tensioning and linking before bone resection in both flexion and extension. This achieves the desired equal rectangular extension and flexion gaps without soft tissue injury. Also, the LATER technique does not align the limb to a predetermined angular measurement as dictated by the MA-TKA, nor is it based on measured resections as described in the KA-TKA. Instead, the goal of the LATER technique is to create a neutral tibial mechanical axis and then build the femoral resection off the individual patient’s soft tissue envelope. This novel method may be a solution to address the findings of Bellemans and colleagues who reported the “normal” limb alignment of 500 healthy male and female subjects ranged from 9 degrees of varus to 9 degrees of valgus in a Bell-shaped distribution (Fig. 7B).16

Similar to Kizaki et al17 who conducted a recent meta-analysis looking at almost 4000 patient specific-TKAs which found that there was a decrease in blood loss secondary to the avoidance of intramedullary violation, we too concluded that there was decreased blood loss with the extramedullary technique. We also theorize that the lack of soft tissue releases may have prevented additional blood loss and contributed to less swelling and potentially earlier range of motion. Another advantage of the extramedullary technique was that retained hardware was not an issue in our cohort (Figs. 8A, B).

FIGURE 8:

A, Preoperative anteroposterior and lateral radiographs of a patient with posttraumatic osteoarthritis with retained hardware after ACL reconstruction. B, Postoperative imaging of the knee with retained femoral hardware secondary to an extramedullary femoral resection. ACL indicates anterior cruciate ligament.

The lack of soft tissue releases may contribute to the superior patient satisfaction rates. Our findings are similar to those reported for hip arthroplasty by Hamilton et al18 who found that the satisfaction rate for THA was 91% when compared with TKA at 81%. Our satisfaction rate of 89.2% was superior to the historically reported TKA rate of 81%.19 Equally, our patient reported improvement rate of 92.3% is encouraging. Preoperative patient expectations, persistent knee pain, and decreased activity are among the leading causes of patient dissatisfaction after a TKA.4,19,20 We postulate that our cohort patients’ knees felt like their native knees and performed in a more physiological manner because the soft tissue was less traumatized with the LATER technique. This method may be returning the knee to the prearthritic alignment (Fig. 9).

FIGURE 9:

A preoperative long-leg radiograph showing an hip, knee, ankle angle of 12 degrees of varus. Postoperative imaging shows that the new limb alignment using the linked and tensioned extramedullary resections (LATER) technique for total knee arthroplasty has brought the limb back to the same alignment of the nonarthritic knee, 5 degrees of varus. Level of improvement: 4; patient satisfaction: 6.

There are some limitations to our study. First, there is no direct comparison made to another cohort group who underwent either MA-TKA or KA-TKA. Unfortunately, the PI has evolved the LATER technique over several years and the quality of data available before contemporary electronic health records is inadequate for us to present a matched group for comparison. We encourage other surgeons to perform knee replacement surgery with the LATER technique and develop additional research data.

Next, the measurement of radiographic alignment via the HKA angle can be influenced by several factors such as internal and ER of the limb and patient position.21 Malrotation of the limb during long-leg radiographs could lead to falsely under or over-measurement of the varus or valgus alignment of the limb.22,23 Fortunately, the same radiology technologist performed all of the imaging and followed a protocol attempting to keep all of the patients’ feet pointing forward (neutral internal rotation and ER).

Some surgeons may consider another limitation of the LATER technique is not knowing the precise numbers related to the various resections and angles for distal femoral implant positioning. We are comfortable not knowing these numbers because one of the goals of the LATER technique is to create a balanced knee every time, which we believe has contributed to improved patient satisfaction.

Next, there may be concern that this technique will not work in the severely malaligned knee with a fixed deformity. This is the precise reason such a challenging cohort of malaligned knees was identified and found to have similar outcomes as a normal alignment subgroup.

We believe that the only limitation to using the LATER technique is the absence of either or both collateral ligaments. Equally, we did not have an opportunity to perform this LATER technique in a patient with a severe flexion contracture.

Finally, there were a handful of patellar issues postoperatively. Although this study did not research the specific amount of ER on the femoral component via computed tomography scan, a balanced rectangular flexion gap was created based off of the resected tibia. We attribute these failures to the PI transiently switching the suture used to repair the arthrotomy during the reported time frame from a running looped 0 PDS to a newly released barbed suture that was FDA-approved for fascial closure. An examination of all TKA patients who underwent surgery by the PI from 2011 until the cohort group in 2016 and since late-2018 showed no medial arthrotomy disruptions. This window of issues can be attributed to a change in the medial arthrotomy repair, not the technique which has been used for several years prior and has shown no re-emergence after switching back to the original closure technique.

Future studies should aim to determine the distal femoral implant rotation with the LATER technique; cost-effectiveness of the Missing Link Device compared with navigation, robotics, and preoperative scans such as magnetic resonance imaging or computed tomography; preoperative versus postoperative limb alignment in patients who undergo TKA for isolate patellofemoral arthritis; and, finally, postoperative limb alignment compared with a nonarthritic contralateral knee as seen in Figure 9. It is possible that this technique may be closer to achieving the “forgotten knee” as described by Eymard and colleagues.24,25

CONCLUSION

The new LATER technique has been shown to improve coronal alignment following a complex primary TKA. We have also found that the knee can achieve coronal balance without ligament releases. The method’s benefits of avoiding intramedullary violation have shown decreased intraoperative trauma and blood loss. With no special preoperative scans (computed tomography/magnetic resonance imaging) and avoiding the complexities of navigation, the LATER technique of coronally tensioned alignment is simple to use and easily reproducible

ACKNOWLEDGMENTS

Special thanks to Stamatis N. Kantartzis, MD for reviewing the radiographs. The authors thank Amir Jamali, MD and James Shaw, MD for proof-reading the manuscript. Finally, a special appreciation to Melissa Miller for obtaining the patient-reported outcomes data.

REFERENCES 1. Charnley J. Total hip replacement by low-friction arthroplasty. Clin Orthop Relat Res. 1970;72:7–21. 2. Bourne RB, Chesworth BM, Davis AM, et al. Patient satisfaction after total knee arthroplasty: who is satisfied and who is not? Clin Orthop Relat Res. 2009;468:57–63. 3. Clement ND, Patrick-Patel RS, Macdonald D, et al. Total hip replacement: increasing femoral offset improves functional outcome. Arch Orthop Trauma Surg. 2016;136:1317–1323. 4. Noble PC, Conditt MA, Cook KF, et al. The John Insall Award: patient expectations affect satisfaction with total knee arthroplasty. Clin Orthop Relat Res. 2006;452:35–43. 5. Rajgopal A, Dahiya V, Vasdev A, et al. Long-term results of total knee arthroplasty for valgus knees: soft-tissue release technique and implant selection. J Orthop Surg. 2011;19:60–63. 6. Roth JD, Howell SM Matsuda S, Lustig S, van der Merwe W. Soft tissue balance of the native knee provides guidance for balancing a total knee arthroplasty. Soft Tissue Balancing in Total Knee Arthroplasty, (Vol. 1). Berlin, Germany: Springer, Berlin, Heidelberg; 2017:17–27. 7. Rivière C, Iranpour F, Auvinet E, et al. Alignment options for total knee arthroplasty: a systematic review. Orthop Traumatol Surg Res. 2017;103:1047–1056. 8. Abdel MP, Oussedik S, Parratte S, et al. Coronal alignment in total knee replacement. J Bone Joint. 2014;96-B7:857–862. 9. Kahlenberg CA, Nwachukwu BU, Mclawhorn AS, et al. Patient satisfaction after total knee replacement: a systematic review. HSS J. 2018;14:192–201. 10. Beer JD, Petruccelli D, Adili A, et al. Patient perspective survey of total hip vs total knee arthroplasty surgery. J Arthroplasty. 2012;27:865–869. 11. Hamilton D, Henderson GR, Gaston P, et al. Comparative outcomes of total hip and knee arthroplasty: a prospective cohort study. Postgrad Med J. 2012;88:627–631. 12. Scott CEH, Bugler KE, Clement ND, et al. Patient expectations of arthroplasty of the hip and knee. J Bone Joint Surg Br. 2012;94-B:974–981. 13. Cooke D, Scudamore A, Li J, et al. Axial lower-limb alignment: comparison of knee geometry in normal volunteers and osteoarthritis patients. Osteoarthritis Cartilage. 1997;5:39–47. 14. Thiele K, Perka C, Matziolis G, et al. Current failure mechanisms after knee arthroplasty have changed: polyethylene wear is less common in revision surgery. J Bone Joint Surg Am. 2015;97:715–720. 15. Barad SJ, Howell SM, Tom J. Is a shortened length of stay and increased rate of discharge to home associated with a low readmission rate and cost-effectiveness after primary total knee arthroplasty? Arthroplast Today. 2018;4:107–112. 16. Bellemans J, Colyn W, Vandenneucker H, et al. The Chitranjan Ranawat Award: is neutral mechanical alignment normal for all patients? The concept of constitutional varus. Clin Orthop RelatRes. 2011;470:45–53. 17. Kizaki K, Shanmugaraj A, Yamashita F, et al. Total knee arthroplasty using patient-specific instrumentation for osteoarthritis of the knee: a meta-analysis. BMC Musculoskelet Disord. 2019;20:1–8. 18. Hamilton DF, Lane JV, Gaston P, et al. What determines patient satisfaction with surgery? A prospective cohort study of 4709 patients following total joint replacement. BMJ Open. 2013;3:1–7. 19. Kim TK, Chang CB, Kang YG, et al. Causes and predictors of patients dissatisfaction after uncomplicated total knee arthroplasty. J Arthroplasty. 2009;24:263–271. 20. Gunaratne R, Pratt DN, Banda J, et al. Patient dissatisfaction following total knee arthroplasty: a systematic review of the literature. J Arthroplasty. 2017;32:3854–3860. 21. Swanson KE, Stocks GW, Warren PD, et al. Does axial limb rotation affect the alignment measurements in deformed limbs? Clin Orthop Relat Res. 2000;371:246–252. 22. Maderbacher G, Baier C, Benditz A, et al. Presence of rotational errors in long leg radiographs after total knee arthroplasty and impact on measured lower limb and component alignment. Int Orthop. 2017;41:1553–1560. 23. Jamali AA, Meehan JP, Moroski NM, et al. Do small changes in rotation affect measurements of lower extremity limb alignment? J Orthop Surg Res. 2017;12:1–8. 24. Eymard F, Charles-Nelson A, Katsahian S, et al. “Forgotten knee” after total knee replacement: a pragmatic study from a single-centre cohort. Joint Bone Spine. 2015;82:177–181. 25. Eymard F, Charles-Nelson A, Katsahian S, et al. Predictive factors of “Forgotten Knee” acquisition after total knee arthroplasty: long-term follow-up of a large prospective cohort. J Arthroplasty. 2017;32:413–418.